Digitized by the Internet Archive
in 2011 with funding from
University of Toronto
http://www.archive.org/details/illuminatingengi08illu
Transactions
of THE
Illuminating Engineering
Society
vol. VIII
JANUARY- DECEMBER
1913
Subject Index and Index to Authors
ILLUMINATING ENGINEERING SOCIETY
29 WEST THIRTY- NINTH STREET
NEW YORK
/
loo
M 33
"><?
SUBJECT INDEX.
* Pages following the numbers given should be consulted in referring
to a topic or subject.
Absorption of light : page
Aluminum finished reflectors 281, 290
Prismatic and opal glassware 452
Reflectors past and present 107
Accidents :
Caused by poor lighting 139, 474
Attributed to darkness and insufficient light 134
Prevention by proper illumination of streets 91
Acetylene :
Lighting status 333
Spectrum 465
Acuity. (See visual acuity.)
Agricultural lighting, low-voltage lamps and systems 327
Aluminum reflectors 269, 271, 282
Absorption of light 281, 290
Altar lighting 617
Alternating current low frequency for flame carbon. arc lamps 328
Antique and curio shop lighting 524
Architect :
Need of co-operation between decorator and illuminating
engineer 153
And illuminating engineering 344
Arc lamps: (See also flame carbon arc.)
Device for converting enclosed alternating and direct current
carbon arc lamps into flame arc lamps 327
Enclosed carbon, operation characteristics compared with flame
carbon arc lamps 166
Luminous for street lighting 88
Magnetic blow 165
Magnetite arc, increased use 327
Modification of light from carbon arcs to approximate daylight 352
Non-magnetic arcs 165
Peculiar phenomenon of positive -crater 328
Stability of arc influenced by wind, magnetic fields, etc 328
■ Status of open and enclosed series and multiple arc lamps.... 327
Street lighting installation of New York 202
Theory of luminosity 407
Titanium arc characteristics 407
Ultra-violet light radiation 165
_ Variations in length of arc with voltage changes 165
vs. tungsten lamps in department store lighting 17
IV TRANS. I. E. S. VOL. VIII
PAGE
Automobile show lighting (gas) 265
Barber shop, lighting 508
Bedroom lighting 259
Book store, lighting 524
Brightness: (See also intrinsic brilliancy.)
A new instrument for observing intrinsic brilliancy 341
Flame arc lamps 483
Of tungsten lamp filaments 353
Of various colored surfaces exposed to different illuminants . . 70
Of the sky 233
Test for discrimination 43
Cadmium-vapor tube lamp, efficiency 329
Calculation :
A method for determining illumination at any point on a flat
surface 355
Candle-power: (See also Deterioration.)
Cause of falling off in mantles 332
Deterioration in gas lighting units attributed to mantle burner
and glassware 332
Of high pressure gas lamps 331
Low pressure high efficiency gas lamps for exterior lighting . . 330
Tests of tungsten lamps 550
Variation in electric incandescent lamps produced by bulb
frosting 326
Variation in electric incandescent lamps with voltage change 595
Candy shop, lighting 520
Carbons, arc. (See electrodes.)
Carbon electric incandescent lamps :
Change in candle-power with variation in voltage 595
Damage in show-case lighting 2>7
-, — Method of testing for spots in filaments 326
Percentage of total incandescent lamp sales for three months
of 1913 656
Relative number of carbon, gem, tantalum electric incandescent
lamps sold in 1907, 1908, 1909, 1910, 191 1, 1912 323
versus tungsten lamps for seeing 146
Carcel lamp, color of light 316
Car lighting. (See railway car lighting.)
Carpet store 508
Cataract of the eye caused by injurious light 132
Chancel lighting 617
Chromatic aberration of the eye 298
Church lighting :
Altar 617'
Apparatus and fixtures 618
Average intensity of illumination 624
SUBJECT INDEX ' • V
Church lighting {continued) :
Bulletin board lighting 622
Chancel lighting 617
Deck system 614, 625, 627, 628
Exterior illumination 618
General requirements 617
Use of indirect lighting 615, 616, 623, 626
Color :
Artificial daylight booth for matching colors 34, 38
Comparative diffusiveness of lights of different colo; 232
Light for theatrical effects 35
Loss of efficiency of the eye caused by lights of different color 256
— In lighting installations 663
— Limen test, for color sensitivity of the eye 41
— The measurement of lights of different color 302
Perception by the eye 343
— Psychic value of illumination 357
Visibility of railway signal lights of various colors 299
Colorimeter, use and difference from spectro-photometer 71
Colored surroundings, influence on color of useful light 62
Coefficients, reflection 68
Costs :
Artificial daylight for interior illumination 253
Department store lighting ^2
Industrial lighting 476, 486, 672
Railway car lighting 602
Small home lighting 672
Cove lighting 615
Deficiencies 114
Daylight :
An approximation of an interior distribution with artificial
light sources 244
An example of a copy of daylight with artificial light sources.. 234
Characteristics 233, 238
Compared with artificial for interior illumination 134
Compared with indirect lighting 626
Cost of artificial daylight 253
Desirable qualities found in window lighting by daylight 252
Diffusion of daylight compared with artificial light sources... 232
Direction and distribution of daylight in interiors 113
Efficiency of tungsten lamp approximating the color of day-
light 326
The ideal illuminant 404
Intensities compared with interior illumination intensities.... 112
' Intrinsic brilliancy of the sky 233
vi TRANS. I. K. S. VOL. VIII
PAGE
Daylight [continued) :
Modification of light from electric arc lamps to approximate
daylight 352
— — Simulation for interior illumination 124
Definitions. (See proposed definitions, page 6, part 1, June (1913)
issue.)
Delicacies shop, lighting 528
Department store lighting:
— Artificial daylight for color matching 34
Avoidance of high intrinsic brilliancy 19
Cleaning lighting equipment 20
-. Color of light required 19
Costs of cleaning and maintenance 32
Deterioration 37
Fixtures (direct) 21
Important considerations involved 19
Lighting systems 17
McCreery & Co., Pittsburgh, store 17
Requirements 19, 33, 36
Rug racks, lighting 22
Show cases 30
Show windows (see subject : Window lighting) 28
Deterioration :
— i — Illumination in a department store 37
— Of light from aluminum finished reflectors 281, 290
- In gas lighting units attributed to mantle burner and glass-
ware 332
Cause of candle-power deterioration of mantles 332
Diffuse reflection :
Explained and illustrated 270
Diffusion :
Its importance in illumination 123, 125, 661
Daylight compared with artificial light sources 232
Dining room lighting 259
Direct lighting :
Compared with indirect and semi-indirect 116
Compared with indirect lighting in a department store 18
Abuses in application 114
Comparative illuminations produced from sources of the same
flux by direct, indirect and semi-indirect methods 118
In churches .- 614
Efficiency :
Basis for comparing reflectors 280
Cadmium-vapor tube lamp 329,
Daylight approximation with use of tungsten lamp 326
SUBJECT INDEX Vli
PAGE
Efficiency {continued):
Eye after a period of work 45
Eye under lights of various colors 145
Flame arc lamps 328, 485
Good lighting as an essential, in factories 137
Loss of efficiency of the eye under lights of different color... 256
Low pressure high efficiency gas lamps for exterior lighting.. 330
Of mercury-vapor lamp combined with tungsten lamp light
to give white light 328
Of Moore dioxid lamp 376
Of Neon tube lamp 376
Of nitrogen, filled tungsten lamp 325
Of various types of reflectors 289
Tests of the eye under different systems of lighting 41
Of various tungsten lamps 324
Quartz tube lamp for street lighting 329
Electrodes :
Flame electrodes. (See Flame carbon arc lamps.)
Enclosed carbon operation characteristics compared with flame
carbon arc lamps 166
Peculiar phenomenon of positive crater 328
Titanium arc characteristics 407
■ Ultra-violet light radiation '. 165
Electron theory, mercury-vapor lamps 77, 86
Eye: (See also Visual acuity.)
■ Angle of direction of light, its effect on the efficiency of
the eye 277
Bottle makers' cataract and an explanation of its cause 132
Causes of discomfort 54
Chromatic aberration 298
Discomfort, indications, caused by illumination 53
Discomfort and influence caused by direction of light 255
Effect upon the eye of ultra-violet light emitted by commer-
cial light sources 343
Efficiency 256
Efficiency under lights of various color 145
— • Efficiency under several systems of illumination 256, 257
Elimination of glare — a cause of .ocular discomfort no
r Eyestrain ; Its nature and cause 141
— Fatigue and eyestrain caused by watching motion pic-
tures 188, 190, 191
Influence of ultra-violet light 132, 138
Loss of efficiency after a period of work 45, 51
Loss of efficiency under lights of different color 256
Perception of color 343
V1I1 TRANS. I. E. S. VOL. VIII
PAGE
Eye {continued) :
Research to ascertain effect of luminous radiation upon the eye 342
Tests for brightness discrimination 43
Tests for color sensitivity 42
- Tests for efficiency and discomfort under different systems of
illumination 40, 58
Tests for visual acuity 43
- Vision as influenced by the brightness of surroundings
292, 297, 298, 299
Walls : An element affecting ocular comfort 108
Eyestrain :
And illumination in industrial buildings 277
And illumination 128, 130
And glare from tungsten lamps in railway cars 608, 609
Exhibition hall lighting:
Semi-indirect (gas) 263
Factory lighting: (See also Industrial lighting.)
Progress and status 486
Costs 476, 486, 672
Average illumination intensity 658
Firefly :
■ Intrinsic brilliancy 335
Fixtures :
A novel fixture in a state building 351
■ And apparatus for church lighting 618
Department store (direct) 21
Distinctive store lighting 522
Hospital 489
Poor illumination of interiors attributed to fixture manu-
facturers 154
Railway car 214
Semi-indirect 264
Substitution of marble for glassware 352
■ Window lighting 564
Flame carbon arc lamps :
— — Applications 327
Brilliancy, intrinsic 483
Candle-power variation with voltage and current variations.. 171
Comparative light giving efficiency of yellow and white carbons 175
Candle-power and operation characteristics 177, 328
Color of light depends upon current density 164
Composition, color of light, and operation characteristics 168
Efficiency of yellow and white carbons for industrial lighting 484
Factors which determine light X73
SUBJECT INDEX ix
PAGE
Flame arc lamps {continued):
Improvement in carbons 327
Life of carbons 327
Scheme for determining amount of illuminants in an electrode 170
Comparative performance on direct and alternating current
circuits 174
Current densities of electrodes and terminals 167
Efficiency 328, 485
Device for converting enclosed alternating and direct current
lamps into a flame arc lamp 327
Displacing open and enclosed arc lamps 327
For low frequency circuits 328
Controversy with high pressure gas lighting people over a
street installation in England 349
Heights for street installations 485
Improvements in construction 327
Magnetic lamps 165
Light independent of polarity 328
Maintenance 176
Operation characteristics 163, 166
Peculiar phenomenon of positive crater 328
Glassware .' 175
Globes for prevention of soot 172
Slagging 171
Stability in flame carbon arc lamps 165, 328
Advantages for street lighting 175
Street lighting in Chicago 485
Street lighting in New York 202
Variations in length of arc with voltage changes 165
Theory of luminosity 407
Ultra-violet light radiation 165
Focusing tungsten lamp, construction and efficiency 326
Focusing type reflectors 274
Gas lamps :
Cause of falling off of candle-power 332
Characteristics of a high power single mantle inverted lamp.. 331
Candle-power deterioration in gas lighting units attributed to
mantle burner and glassware 332
High pressure. (See High pressure gas lighting.)
Modification of light to approximate daylight 352
New types of units 330
Number used in street illumination in New York 202
Artificial silk mantle 332
Theory of luminosity of Welsbach mantle 407
X TRANS. I. K. S. VOL. VIII
PAGE
Gas lighting: (See also gas lamps.)
Acetylene 333
Cause of falling off of candle-power 332
Golf courts 347
High pressure lamps, difficulties and improvements 331
Ignition, elements used 332
Low pressure high efficiency lamps for exterior lighting 330
Candle-power and efficiency 331
Natural gas, a new method of utilization 333
New types of units . , 330
Gas pressure in several cities of United States 5 13
Semi-indirect in an exhibition hall 263
In small stores 499
In a Sunday-school room 439
Tennis courts 347
Gem lamps :
Relative number of carbon, gem, tantalum electric incandes-
cent lamps sold in 1907, 1908, 1909, 1910, 191 1, 1912 323
Percentage of total incandescent lamp sales for three months
of 1913 656
Glare :
Cause of drowsiness during church services 624
Elimination by location of light sources no
Influence on seeing 104, 109
From lamps 608
Means of elimination no
Its nature and character , 667
Patented bulb for reducing glare of electric lamp filament.... 327
From tungsten lamps in railway cars 608, 609
Glassware :
Characteristics (distribution, absorption, color, transmission)
of prismatic and various enclosing glassware 447
Distribution 452
Effect upon life of tungsten lamps 456
For flame carbon arc lamps 175
Globes for the prevention of soot accumulation from flame
carbon arc lamps 172
Substitution of marble for glassware on lighting fixtures.... 352
Golf courts, lighting '. 347
Great white way lighting 90
Grocery store, lighting , 521
Hefner lamp as a standard in photometry 412, 422'
Heterochromatic photometry, a practical solution 302
SUBJECT INDEX xi
PAGE
High pressure gas lighting:
Candle-power of lamps 33 1
Controversy with electric people over installation in England . . 349
Difficulties and improvements in lamps 331
Progress and status of street lighting 350
Street lighting status, United States and abroad 331
Home lighting :
Advantages and disadvantages of home experiments 230
Experiments in illumination from large area light sources.... 229
Influence of light and dark walls upon the illumination
required 151
Location of light sources 126
Requirements 151
Simulation of daylight for interior illumination 124
Hospital lighting:
Corridors 498
Operating rooms 488, 498
Operating table lamp 493
Wards 494, 498
Zeiss system, illumination intensities ■ 497
Illuminating engineer :
His work and qualifications 100, 169, 669
Need of co-operation with architect 153
Illuminating engineering :
As a science and an art 100
And architects 345
Use of photography 554
Laboratory and equipment of General Electric Co 379
Illuminating Engineering Society:
Classification of papers 6
Committees activities ( 1912) 8
Committee reports and activities 689
Finances 683
Functions 3
General Secretary's report .' 6, 683
History and growth 1
Illumination primer "Light: Its Use and Misuse" 675
Membership classification 14, 684
Its unique position among societies 676
Its relation to public lighting companies 99
Report of General Secretary 6
Sections 687
Transactions 688
xii TRANS. I. E. S. VOL. VIII
PAGE
Illuminants : (See also lamps according to name.)
Daylight, the ideal illuminant 4°4
Determination of radiation efficiency 404
Radiation efficiency 404
Ultra-violet radiation from artificial light sources 132
Use of spectrum in determining value 407
Illumination: (See also lighting.)
Angle of direction of light for efficiency of the eye 277
Average intensities of store, street, show window, factory,
office, residence, railway cars 658
Average intensity in churches 624
Comparative illuminations produced from sources of the same
flux by direct, indirect and semi-indirect methods 118
Concerning data on competitive illuminants 514
Contrast in illumination; its nature and importance 666
Department store intensities 31
Direction and distribution of daylight in interiors 113
Elimination of streaky effect from metal reflectors 286
And eyestrain 128, 130, 277
Inadequacy of measurements in the horizontal plane 158-160
Increased industrial production with good illumination 473, 485
Increases in intensities in recent years 655
And industrial accidents 474
Influence of colored surroundings on the color of the useful
light 61, 107
Influence of decoration and contrast in the illumination of
interiors .■ 108
Influence of light between 60° and horizontal upon sight 291
Intensities. (See Illumination intensities.)
Intensity required for street lighting 89
Interiors, various phases and problems 99
Legislative requirements in Holland, England, New York and
Wisconsin 353
A method for determining illumination at any point on a flat
surface 355
A method for plotting distribution curves , 355
Its principal use 101
Progress during the year ... 323
Psychology of intensities 284
Relative intensity of sunlight and artificial illumination 301
Reflectors. (See Reflectors.)
Tests for the efficiency of the eye under different systems of
lighting 40
Typical street lighting intensities 658
SUBJECT INDEX xiii
T J , PAGE
Incandescent electric lamps: (See also lamps by name.)
Average candle-power of all incandescent lamps sold 1906-
«W3 657
Cooling effect of leading-in wires upon lamps of the street
series type 3g5
Helical filaments 537
Method of testing for spots in incandescent lamps filaments... 326
Relative number of carbon, gem, tantalum and tungsten
electric incandescent lamps sold in 1907, 1908, 1909, 1910,
191 1, 1912 32,
A new method of determining the true temperature of fila-
ments *,->!
334
Patented bulb for reducing glare of filament 327
Recent improvements in manufacture 534
Smashing point S40) 546, 547
Sign lamps ,24
Indirect lighting:
Comparative illumination produced from sources of the same
flux by direct, indirect and semi-indirect methods 118
Compared with daylight 626
Compared with direct and semi-indirect ' 116
Compared with indirect in a department store 18
Fixtures for railway car lighting 348
Hospital fixture 492
In churches 615, 616, 623, 626
Its influence upon the design of lighting installations 114
Industrial lighting:
Accidents caused by poor illumination and darkness .. 134, 139, 474
Costs 476, 486, 672
Good lighting as an essential to efficiency 137
In factories 486
Influence of good lighting on production 287
Legislative requirements in Holland, England, New York and
Wisconsin 353
Mercury-vapor lamps 479
Metal reflectors 268
Progress 345, 486
Percentage increase in output with good illumination 473, 485
Standards compared with store illumination 283
Total annual cost 487
Intensities of illumination :
Average in churches 624
Department store illumination 32
Factory lighting 658
Motion picture screens 195
Increases in recent years 655
XIV TRANS. I. E. S. VOL. VIII
PAGE
Intensity of illumination {continued):
Industrial lighting 658
Office lighting 658
Railroad car lighting 223, 658
Residence lighting 658
Show window lighting 658
Store lighting 658
Street lighting 658
Intensive type reflector 274
Interior illumination :
Requirements 101
Various phases discussed and illustrated 99
The window as a light source 241
Intrinsic brilliancy :
Of firefly 335
Flame carbon arc lamps 483
Of the sky 233
Tungsten lamp filament 233, 353, 483
Welsbach mantle 233
Jewelry store lighting 508, 517
Ladies' wear shop lighting. 525
Lamps. (See lamps according to name, arc, incandescent, gas, etc.)
Kerosene lamp : A reason for its present day use 132
Miniature tungsten lamps, construction details 325
Carbon arc 162
Relative number of carbon gem, tantalum, and tungsten elec-
tric incandescent lamps sold in 1907, 1908, 1909, 1910,
1911, 1912 323
Lamp-posts :
New York street lighting 200
Ornamental for street lighting 92
Legislation, lighting 347
Industrial illumination requirements in Holland, England, New
York, and Wisconsin 353
Law of reflection 270
Library lighting :
Investigation conducted by British Illuminating Engineering
Society 344
Light :
"Cold" light 334
Color best adapted to seeing 143
Color of light required for motion picture projectors 184
SUBJECT INDEX XV
PAGE
Light {continued):
Direction of light in interiors 659
Discomfort and influence on the eye caused by direction of
light 255
Explanation and theory 131
Kerosene light for seeing 132
Perception of lights of short duration 342
Poor illumination attributable to fixture manufacturers 154
Psychic value of light, shape, form and color 357
Radiation of carbon incandescent lamps and tungsten lamps
compared for seeing 146
Some theoretical considerations of production 400
Various colors and their influence upon vision 145
Yellow versus white light for seeing 145
Lighting: (See also indirect and semi-indirect lighting; illumination.)
Abuses in application of direct lighting 114
Agricultural, low voltage lamp and system 327
Angle of direction of light for efficiency of the eye 277
Approximation of an interior daylight distribution with arti-
ficial light sources 244
Artificial window lighting 250
A cause for industrial accidents 139
Bedroom 259
Causes of poor illumination 275
Characteristics of daylight illumination 233,238
Church 613
A classification of light sources 236
Color ; its use in lighting 663
Colored surroundings, influence on the color of the useful light 61
Comparative illuminations produced from sources of the same
flux by direct, indirect and semi-indirect methods 118
Concerning data on competitive illuminants 514
Congruity in design 668
Costs for lighting of various interiors 672
Cove lighting deficiencies 1 14
Daylight : the ideal illuminant 404
Department store (see heading: Department store lighting for
sub-divisions) '. 17
Diffusion — growing importance 661
Dining-room 259
Direction of light for interior illumination. .124, 126, 136, 260, 659
Direction and distribution of daylight in interiors 113
Direct lighting compared with indirect and semi-indirect 116
Factories 470
Gas lighting. (See topics of gas lighting.)
XVI TRANS. I. E. S. VOI,. VIII
Lighting {continued): page
Of home 149
Home experiments in illumination from large area light
sources 229
Hospitals 488
Hygiene and safety as influenced by lighting 670
Indirect compared with direct in a department store 18
Industrial 470
Industrial establishments 672
Influence of decoration and contrast in the illumination of
interiors 108
Influence of good lighting upon industrial production 136, 287
Importance of good lighting 368
Influence of light between 6o° and horizontal upon sight 291
Influence of surroundings 107
Intensities of interior illumination compared with daylight
intensities 112
Interiors, various phases and problems 09
Legislation 347
Library investigation by British Illuminating Engineering
Society 344
Location of light sources for home illumination 126
Motion picture projectors 180
Moving picture screens and auditoriums 187
Policies on lighting installations of different companies
481, 483, 5i6
Its principal use 101
A psychological aspect 284
Progress 323, 345
Railroad passenger cars 214
Railway cars 597
Rug racks 22
Rural districts, sundry systems 333
Semi-indirect from side walls 246, 261
Simulation of daylight for interior illumination 124
Standards compared with store illumination 283
Status of lighting arc 652
Steadiness of light, its importance 665
Store. (See Store lighting.)
Street. (See topic street lighting.)
- A Sunday-school room (gas) 439
Tennis courts (by gas) 347
Tests for the efficiency of the eye under different systems of
illumination 40,
Typical intensities, store, street, show windows, factories,
offices, residences, railway cars 658
SUBJECT INDEX xvii
PAGE
Lighting Companies :
Progressive attitude in lighting improvements 676
Limen test for color sensitivity 41
Luminosity :
Theory for various illuminants 407
Lunch room lighting 508
Magnetic blow in arc lamps illustrated 165
Magnetite arc lamps increased use 327
Mantles : (Gas.) 332
Cause of falling off of candle-power 332
Developments in silk mantles 332
Theory of luminosity of Welsbach mantle 407
Mercantile establishment lighting 672
Mazda lamps. (See tungsten lamps.)
Mercury-vapor lamps :
"Reluctance" explained 76, 86
Theory, operating and starting characteristics 75, 86
Voltage required depends upon vapor pressure 76
Disappearance of gas due to chemical action rather than physi-
cal absorption ; 330
Developments in sealing 329
Efficiency of mercury-vapor lamp combined with tungsten
lamp, light to give white light 328
Characteristics of quartz burner 84
In industrial lighting 479, 487
Millinery store lighting 519
Moore dioxid tube lamp efficiency 376
Motion picture lighting:
Candle-power performance of projectors on alternating and
direct current 185
Color of light required 184, 195, 198
Fatigue and eye strain 185
Intensity of illumination on screens 194
Illumination of projectors 180
Mechanism of lamps 182
Projection screens: Advantages and disadvantages 186
Nomenclature and standards. (See proposed definitions, page 6, part
1, June issue.)
Nitrogen filled tungsten lamp 325
"Movies." (See motion pictures.)
Neon tube lamp characteristics 330
Short life due to absorption of neon by electrodes 330
Characteristics and performance of lamps 371
XV111 TRANS. I. E. S. VOL. VIII
PAGE
New York City:
Street lighting equipment 199
Office lighting :
Average illumination intensity 658
Parkway lighting 90, 208
Passenger car lighting: (See railing car lighting.)
Pentane lamp :
Accuracy values in photometry 419, 435
Adjustment of lamps 415
Candle-power variations and peculiarities 434
Details of operation 416
Effect of pentane density on candle-power of pentane lamps.. 425
Effect of atmospheric conditions 427
General directions for use 420
Standardization at the United States Bureau of Standards... 413
Uses, characteristics and deficiencies 41 1
Variation of candle-power with height of flame 417
Variation of candle-power with lumidity and barometric pres-
sure 429
Variations in candle-power with different grades of pentane . . 436
Photo-electric cell :
In photometry 459
Sensitivity 320, 468
History 460
Limitations and objections 465
Methods of use 460
Sensibility curve 464
Photography :
In illuminating engineering 354
Photometric curves :
Intensive type reflector 274
Distributing type reflectors 274
Extensive type reflector 271
Focusing type reflector 274
Photometry :
Accuracy in studies 633, 648, 678
Accuracy of Bunsen and Lummer-Brodhun devices 634
Average error of test plates 643
Accuracy variation of photometer operators 647
Accuracy values of pentane lamp 419
Colorimeter, use 71
Colored lights, measurement. (See heterochromatic below.)
Description of new photometers 339,
Deficiencies of hemispherical and other test screens 158
Error due to parallelism of rays of reflected light sources 635
Primary standard of light, progress of investigation ....... . . .' 335
- Inadequacy of measurements in horizontal plane i58; 161
SUBJECT INDEX xjx
Photometry (continued): PAGE
Errors in illumination measurements due to failure of test
plate to obey the cosine law 6
Flicker photometer 4"
Hefner as a standard
Heterochromatic '
Method of reducing the quantity of light "for" photometriS 3
purposes
Method of investigating adjustment errors and' the 'calibration 3'
of portable photometers fi
Method of measuring the energy of ultra-violet "radiation
emitted by mercury-vapor lamp
Pentane lamp as a working standard ., 4I0
Planes for measuring illumination intensities . . . . ' 228
Photo-electric cell, use
Primary standard of lig
Inadequacy of measurei
Single mirror crane type .
Selenium cell, use and characteristics ' .' 33g
Spectro-photometer, use
Street illumination tests with various screens '. I5-
Comparison of tests with screens of various "shapes "and
materials ,
Use of helical filament lamps as standards 545', 548
Pool room lighting -
Post office department, railway car lighting specifications .'.' 346
Prismatic glassware, light absorption and transmission 4S2
Angle of prisms
Distribution \\ .
Primary standard of light; progress of investigation 335
Psychology :
Of illumination fi
Psychic value of light, shade, form and color.........'.'..'.'.'.'.' 357
Quartz tube lamp R
Efficiency
* "Z2Q
Use of ultra-violet rays for destruction of bacteria. . . 320
For street lighting " 32g
Radiation :
Of carcel lamp ,
Efficiency of illuminants '.'.'.'.'. .' 40-
Research to ascertain effect of luminous radiation upon the eye 342
— Selective, explained .403, &
Railroad signals (visibility of signal lights of
various colors) 299
XX TRANS. I. E. S. VOI,. VIII
PAGE
Railway car lighting:
Car lighting specifications of the post office department 346
Center deck system 611
Comparative tests with various luminants 217
Cost data and comparisons 599. 602
Concentrated filament tungsten lamps for locomotive headlights 542
Glare from tungsten lamps 608, 609
Headlights, legislation 347
History 215
Illumination test data 597
— — Indirect fixtures 348
Intensities of illumination 223, 228
Average illumination intensity 658
Lighting schedule 601
Modern practise 589
Methods of producing energy for illumination; their respective
advantages and disadvantages 225
Plans for photometering illumination 228
Progress and status 602
Use of 56-watt and 94-watt tungsten lamps, comparison
592, 604, 606
Wiring 227, 593
Tests in New York subway 606
Reflection :
Classes of reflecting surfaces 270
Coefficients 69
Diffuse 220
Influence of colored surfaces on light reflected 62
Influence of character of reflecting surfaces on the design
and light distribution of reflectors 273
Law of, explained and illustrated 270
Regular 270
Specular 270
Specular influence on light reflected 67
Reflectors: (See also prismatic.)
Aluminum 282
Aluminum and porcelain enameled finishes for reflectors com-
pared 281
Advantages of reflectors of various metals 269
A reflector for copying the distribution of daylight from a
window 248
Absorption of light 107, 281, 290, 452,
Absorption of light by aluminum finished reflectors 290
Basis for comparing efficiencies 28a
Characteristics of various metal reflectors 271
SUBJECT INDEX xxi
Reflectors {continued): page
Considerations which govern selection for an installation 277
Cost considerations in the design of lighting installations 282
Efficiency of several types 289
Elimination of streaky illumination effect from metal reflectors 286
Glass vs. metal for industrial lighting 288
Historical notes on the design of metal reflectors 285
Influence of shape of reflectors on light distribution and
efficiency 289
Influence of deep reflectors on life of electric lamps 288
Influence of character of reflecting surfaces on the design and
light distribution of reflectors 273
Light distribution depending upon design 105
Photometric curve of extensive type reflector 274
Photometric curve of distributing type reflectors 274
Photometric curve of intensive type of reflectors 274
Porcelain enamel, characteristics 282
Proposal to use 1,000 lumen basis in making distribution curves
from reflectors 645, 647
Spacing and mounting heights for various types of reflectors.. 272
Uses and purposes 104
Regular reflection :
Explained and illustrated 270
Residence lighting:
Average illumination intensity 658
Cost in small homes 672
Restaurant lighting 527
Rug rack lighting 22
Selective radiation explained 403, 406
Seeing: (See also visual acuity.)
Influence of glare 104
Efficiency under lights of various colors 144
Selling illumination 499
Semi-indirect lighting :
Comparative illuminations produced from sources of the same
flux by direct, indirect and semi-indirect methods 118
Compared with direct and indirect 116
By gas in an exhibition hall 263
In churches 616
From walls .' 246, 261
Shoe store lighting 508, 522
Show case lighting 30
Show window lighting. (See window lighting.)
Signals :
Visibility of railroad signals of various colors 299
Perception of lights of short duration 342
Use of concentrated filament lamps 542
xxii TRANS. I. E. S. VOL. VIII
PAGE
Sign lamps (tungsten 324, 539
Slagging of flame carbon electrodes ; cause explained 171
Store (small) lighting ^72
Snellen type test for efficiency of the eye 42
Spacings :
And heights for various types of reflectors 272
Spectrum :
Of acetylene 4°5
Use in determination of value of units 4°7
Specular reflection 667
Influence on color of light reflected from colored surfaces 67
— ■— Explained and illustrated 270
Standards :
The pentane lamp in photometry 4xo
The Hefner lamp 412, 422
Primary, progress of investigation of 335
Store lighting:
Antique and curio shop 524
Carpet store 5<>8
Department: (McCreery & Co., Pittsburgh.) See also depart-
ment store lighting 17
Barber shop 508
Books 524
Candy 520
Delicacies 528
Grocery store 521
Jewelry 508, 517
Ladies wear 525
Lunch room 508
Millinery . 519
■ Pool room 5°8
Restaurant 527
Shoe 508, 522
Stationery 526
Standards compared with store illumination 283
— — Tailor-shop 508
Tea room 522
■ Toggery or haberdashery shop 5J8
Toy 517
Street car lighting. (See railway car lighting.)
Street lighting:
Average and typical illumination intensity 658
■ Business streets requirements 88
Color of light required 89
SUBJECT INDEX xxiii
PAGE
Street lighting {continued):
■ Controversy over an installation of high pressure gas and
flaming arc lamps in England 349
Discernment of objects by their silhouettes 94
Flame arc lamps in Chicago 485
Of high pressure gas lighting, progress and status 350
Intensity required 89
Inadequacy of photometrical measurements in horizontal plane 158
New York City 199
Ornamental lamp-posts 92
Height of flame arc lamps 485
Photometry with various screens 155
Progress and status 348
Requirements 88
Spacing and height of illuminants 92, 95, 96
Residence streets requirements 90, 97
Status in France 351
With ornamental arc lamps 88
Sunday school room lighting with gas 439
Sunlight, intensity compared with artificial illumination 301
Tantalum lamps, relative number of carbon gem, tantalum electric
incandescent lamps sold in 1907, 1908, 1909, 1910, 191 1, 1912. 323
Tailor shop lighting • 508
Tea room lighting 522
Tennis court lighting (gas) 347
Tests :
For brightness discrimination 43
For the efficiency of the eye under different systems of illu-
mination 40
Limen test for color sensitivity 41
Snellen type test for the efficiency of the eye 43
Visual acuity 43
Railroad car lighting 217
Theatre lighting :
Color of light for make-ups 35
Titanium arc lamps, characteristics 407
Toy Store 517
Train lighting. (See railway car lighting.)
Trolley car lighting. (See railway car lighting.)
Tungsten lamps :
Blackening of bulbs, cause of 549
Bulb size, decrease 536
Candle-power change with variation in voltage 595
Brightness of filament 233, 353, 483
XXIV TRANS. I. E. S. VOL. VIII
PAGE
Tungsten lamps {continued) :
Characteristics of tungsten filament railway lamps 592
Cooling effect the leading-in wires upon filaments of lamps of
street series types 385
Concentrated filament 325, 537. 54*. 543
Discoloration of lamp bulbs due to chemicals, and its effect
upon candle-power 546
Daylight approximation lamp 326
Effect of frosting on effective light 288
Effect upon life, of enclosing glassware 546
Efficiencies 324
— — Focusing lamp, construction and efficiency 326
Helical filaments 537
Helical filament lamps as standards in photometry 545, 548
Improvements in candle-power performance 534
Influence of deep reflectors on life 288
Intrinsic brilliancy of filament 233, 353, 483
Life and candle-power tests 543, 550, 553
Light of tungsten lamps vs. carbon lamps for seeing 146
Miniature construction details 325
Method of testing for spots in incandescent lamps 326
Patented bulb for reducing glare of filament 327
Percentage of incandescent lamp sales for 3 months of 19 13.. 656
56 and 94-watt railway tungsten lamp systems compared
594, 604, 606
Recent improvements 324, 553
Use of chemicals to prevent bulb blackening 544, 547, 549
Relative number of carbon, gem, tantalum, electric incandes-
cent lamps sold in 1907, 1908, 1909, 1910, 191 1, 1912 223
Reducing blackening of bulbs with "vacuum getter" 324
Sign lamps 324
Smashing point 540, 546, 547
Standardizations 536
Variation in candle-power produced by bulb frosting 326
vs. arc lamps in a department store 17
Ultra-violet light :
"Chinese white" test for presence 355
Effect upon the eye of ultra-violet light emitted by commer-
cial light sources 343
From artificial light sources 132
Influence on the eye 133, 138
Media opaque and transparent to ultra-violet radiation 140
SUBJECT INDEX XXV
PAGE
Ultra-violet light {continued):
- A method of measuring the energy of ultra-violet radiation
emitted by mercury-vapor lamp 340
Radiated from arc lamps 165
Use of rays from quartz tube lamp for destruction of bacteria 329
Vacuum tube lamps :
Cadmium vapor lamp 329
Developments in sealing 329
Efficiency of Neon and Moore tube lamps 376
Disappearance of gas due to chemical action rather than
physical absorption 330
Mercury-vapor 329
Neon tube lamp 330, 376
Vision. (See visual acuity.)
Visual acuity:
And monochromatic light 298
Influenced by brightness of surroundings J92, 297, 298, 299
Influence of light between 60 ° and horizontal upon sight 291
Influence of angle of direction of light upon efficiency of eye.. 277
Tests for the eye 43
Tests of legibility of type 342
Under light of various colors 145
Vision influenced from color of light 143
Visibility of signal lights of various colors 299
Window lighting:
Average illumination intensity 658
Department store 28, 557
Intensity of illumination required 89
Selection and spacing of reflectors 573
Specifications, requirements and tests 557
Walls :
Brightness an element affecting ocular comfort 108
Welsbach mantle :
■ Intrinsic brilliancy 233
Zeiss refractometer for determination for CO- content in atmos-
phere 414, 435
Zeiss system of illumination intensities for hospitals 497
INDEX TO AUTHORS
The letter d indicates discussion.
PAGE
Alger, Ellice M. (M. D.). Illumination and eyestrain 130
d — Eyestrain and motion pictures 190
Amrine, T. H. The cooling effect of leading-in wires upon the fila-
ments of tungsten incandescent lamps of the street series
type 385
d — Accuracy in photometry with Bunsen and Lummer-Brodhun
photometer 647
Baldwin, A. T., W. R. Mott and R. B. Chillas, Jr. d — Flame arc
lamp carbons 177
Barrows, George S. d — Co-operation of architect, decorator and
illuminating engineer in illumination problems 153
Bettcher, C. W. d — Candle-power variation of tungsten lamps due
to change in voltage 609
Bond, C. O. d — Test for the efficiency of the eye 59
d — Home lighting 261
d — Use of reflectors in railroad lighting 284
CalvERT, H. d — Industrial lighting 283
d — Hospital ward lighting 497
d — Central station free renewal policy 543
CauldwEll, F. C. d — Church lighting 625
Chapin, H. C. d — White and yellow flame carbon arc lamps for
factory lighting 484
Chillas, R. B., Jr., W. R. Mott and A. T. Baldwin, d — Flame arc
lamp carbons 177
Claude, George. Neon tube lighting 371
ClEwell, Clarence E. d — Street lighting 212
Cobb, Percy W. Vision as influenced by the brightness of sur-
roundings 292, 299
d — Effect of illumination on the eye 138
d — Illumination intensity and motion pictures 194
Cotton, A. C. d — Illumination of passenger cars 225
CowlES, J. W. d — Street lighting 212
Cravath, J. R. d — Influence of surroundings on vision 297
d — Show window lighting s 587
d — Church lighting 623, 627
Crittenden, E. C, and A. H. Taylor. The pentane lamp as a work-
ing standard 410
Darrah, W. A. The flame carbon arc lamp 162
Some theoretical considerations of light production _400
Dicker, A. O., and M. H. FlExner. Factory lighting 470, 485
INDEX TO AUTHORS xxvii
PAGE
Dunning, H. S. d — Effect of inclosing globes on life of tungsten
lamps 456
d — Candle-power performances of tungsten lamp 546
Edwards, Evan J., and Ward Harrison. Recent improvements in
incandescent lamp manufacture 533
Some studies in accuracy of photometry 633. 649
Edwards, Evan J. d — Temperature characteristics of electric incan-
descent lamp filaments 308
Ely, Robert B. Church lighting 613, 630
d — Home illumination 259
d — Factory lighting 483
d — Store lighting 531
d — Illumination equipment of street cars 610
Fabry, C. H. A practical solution of the problem of heterochromatic
photometry 302
FerreE, C. E. Test for the efficiency of the eye under different sys-
tems of illumination and a preliminary study of the
causes of discomfort 40, 60
d — Illumination and visual acuity 142
d — Illumination in the home from large area light sources 255
Flexner, M. H., and A. O. Dicker. Factory lighting 470, 485
d — Chicago street lighting with flame arc lamps 485
Ford, Arthur H. A photometer screen for use in tests of street
illumination 155
Gage, H. P. d — Surroundings and vision ; monochromatic light ; rail-
road signals 298
Graves, C. B. d — Indirect lighting 128
Halvorson, C. A. B., Jr. Street lighting with ornamental luminous
arc lamps 88
Harrison, Ward, and Evan J. Edwards. Recent improvements in
incandescent lamp manufacture 533, 547
Some studies in accuracy of photometry 633
Harrison, Ward, d — Department store lighting 37
d — Competitive tests of store lighting 514
d — Wiring of street railway cars 610
Hibben, S. G. Modern practise in street railway illumination. .. .589, 610
d — Department lighting 36
d — Influence of enclosing glassware on light distribution 496
d — Store lighting 529
HoadlEy, Geo. A. Indirect illumination on the eye 121
d — Home lighting 254
d — Industrial lighting reflectors 284
Howell, J. W. d — Recent improvements in tungsten incandescent
lamps 549
Hunter, George Leland. Home lighting 149
XXVlii TRANS. I. E. S. VOL. VIII
PAGE
Israel, Joseph D. Annual report of the General Secretary for the
fiscal year ending September 30, 1913 683
Ives. Herbert E. Some home experiments in illumination from large
area light sources 229, 258
d — Tests of visual efficiency 57
d — Lighting interiors 122
d — Influence of surroundings on vision 298
d — Heterochromatic photometry 319
d — Interior illumination - . 369
d — Use of photo-electric cell in photometry 467
d — Uses of the helical filament lamp 544
Jackson, J. B. d — Cost of lighting units for street car service 610
Josselyn, A. E. d — Some causes of poor illumination 120
Kiefer, L. T. d — Showcase lighting 37
Kilmer, William S. Hospital lighting 488
Kingsbury, Edwin F. Experiments in the illumination of a Sunday-
school room with gas 439
Lacombe, C. F. Street lighting of greater New York 199
Lansingh, V. R. Characteristics of enclosing glassware 447
d — Uses of the helical filament lamp t . . . 545
d — Railway car illumination 609
Law, Clarence L., and A. L. Powell. Distinctive store lighting 515
LEE, J. W. d — Glass and metal reflectors for industrial lighting.... 285
Lewinson, L. J. d — Improvement in the candle-power performance
of tungsten lamps from 191 1 to 1913 552
d — Variations in accuracy of photometry with different pho-
tometers 647
Lewis, F. Park (M. D.). The physic values of light, shade, form
and color 357
LitlE, T. J., Jr. d — Pressure of gas lighting for stores 513
d — Church lighting 628
Little, W. F. d — Metal reflectors for industrial lighting 290
d — Hospital lighting 497
d — Standardization of incandescent lamp filament dimensions.. 541
Lloyd, M. G. d— Temperature characteristics of a helical filament
lamp 543
d — Smashing point of tungsten lamps 546
Lloyd, R. L. d— Direction of light for interior illumination. 260
Lloyd, E. W. d — Factory lighting 484
Luckiesh, M. The influence of- colored surroundings on the color
of the useful light 62
d — Measurement of illumination efficiency 159
d — Influence of surroundings on vision 299
d — Temperature and candle-power characteristics of electric -
incandescent lamp filaments 399
INDEX TO AUTHORS XXIX
PAGE
Luckiesh, M. {continued)
d — Church lighting 629
d — Variations in accuracy of photometry 648
McAllister, A. S. d — Direction of light for interior illumination.. 126
Macbeth, Norman, d — Lack of agreement in standard specifications
of incandescent lamps 540
Madgsick, H. H. d — Factory lighting 485
Marks, L. B. d— Daylight illumination 124
d — Influence of surroundings on vision 297
Martin, J. Frank. The illumination of motion picture projectors.. 180
Millar, Preston S. Progress and functions of the Illuminating
Engineering Society (inaugural address) 1
Report of the general secretary for 1912 6
Some phases of the illumination of interiors 99
The status of the lighting art (presidential address) 652
d — Illumination of passenger cars 224
Minick, J. L. Illumination of passenger cars 214, 226
Moore, D. McFarlan. d — Large area light sources 126
Mott, W. R., R. B. Chillas, Jr., and A. T. Baldwin, d — Flame arc
lamp carbons 177
Myers, R. E. d — Improvements in the manufacture of tungsten lamps 544
Nichols, G. B. d — Prevention of poor illumination by co-operation
of illuminating engineer and architect 153
Philbrick, J. E. d — Store lighting 499
Pierce, Robert F. Gas lighting in an exhibition hall 263
d — Store lighting 511
d — Church lighting 625
Porter, L. C. d — Passenger car lighting 227
d — Uses of the concentrated tungsten filament lamp 542
d — Tungsten lamps for railway cars 604
Powell, A. L., and Clarence L. Law. Distinctive store lighting. .515, 531
Powell, A. L. d— Church lighting 623
Reid, H. A. d — Cost and depreciation of tungsten lamps and reflec-
tors for factory lighting 482
Richtmyer, F. K. The photo-electric cell in photometry 459
Rolph, Thomas W. Metal reflectors for industrial lighting. .268. 287. 289
Roosa, G. W. d — Flame carbon lamps for industrial lighting 483
Rose, S. L. E. The illuminating engineering laboratory of the Gen-
eral Electric Company 379
d — Efficiency of six ampere flame arc lamps 484
d — Plotting photometric distribution curves of incandescent
lamps 647
Rows. E. B. d — Interpreting illumination test data 514
d — Tungsten lamp wiring and illumination for railway cars.... 605
d — Church lighting 628
XXX TRANS. I. E. S. VOL. VIII
PAGE
Rowland, A. J. d — Home lighting 260
d — Development of the metal shade for industrial lighting.... 284
Shalling, H. W. Department store lighting 17, 80
d — Department store lighting 38
Sharp, C. H. d — Daylight illumination 123
d — Photometry of street illumination 158
Simon, Edward L. d — Illumination and motion pictures 197
Skiff, W. M. d — Showcase lighting 35
Smith, Louis C. d — Illumination and motion pictures 197
Stark, A. W. d — Interior illumination 128
SticknEy, G. H. d — Department store lighting 33
d— Measurement of illumination 161
d — Concentrated filament lamp 541
d — Railway car illumination 608
d — Church lighting 629
Taylor, A. H., and E. C. Crittenden. The pentane lamp as a work-
ing standard 410
Thomas, Percy H. Theory of mercury-vapor apparatus 75
Tousey, Sinclair, d — Illumination and eyestrain 140
Vaughn, F. A. d — Eyestrain and motion pictures 192
Ware, R. C. d — Poor illumination and bare lamps 120
Wheeler, H. B. d — Hospital lighting 498
The lighting of show windows 555, 588
m}
TRANSACTIONS
OF THE
Illuminating Engineering Society
Published on the 2Sth of each month, except during Ju!y, August, and September, by the
ILLUMINATING ENGINEERING SOCIETY
General Offices: 29 West Thirty-Ninth Street. New York
Vol. VIII
JANUARY. 1913
No. 1
Index for Volume VII.
The index for Volume VII (1912)
of the Transactions will be mailed in
separate form with the February issue
which will be out about the middle of
March.
The New Transactions.
In this issue an attempt has been
made to improve the general make-up
of the Transactions. Dull finished
paper, free from the objectionable glare
which a reader usually encounters in
magazines and books, has been used
throughout. This number also has a
more appropriate cover. Other im-
provements, in the way of better
arrrangement and presentation of sub-
ject matter, more legible type, and the
like, may be expected later. Such
changes will ultimately afford a more
commendable publication which, it is
hoped, will be of greater interest to the
members of the society and more val-
uable as a journal of reference for
libraries. Beginning with the April
issue each number of the Transactions
will be published on the 28th of the
month.
Annual Committee Reports.
Below are given synopses of the
annual reports of committees, which
were presented at a meeting of the
council held in New York, January 10,
I9I3-
finance committee
All the bills for the year were ap-
proved by the committee. A report
on the books of account for 1912 pre-
pared by Wm. J. Struss & Company,
certified public accountants, accom-
panied the committee's report. This re-
port showed a deficit of $640.07 for the
year, and an impairment of surplus
amounting to $1,435-63- The average
expense per member was estimated as
$7.36. The report appears in full else-
where in this issue of the Transactions.
COMMITTEE ON ILLUMINATION
PRIMER
This report is supplementary to the
report of the committee in the June
(1912) Transactions. Ten thousand
primers (Light: Its Use and Misuse)
had been published, and an edition of
5,000 is now on the press. About 5,500
copies have been distributed free by the
society, 4,000 have been sold and deliv-
ered, and orders are on hand for 4,500.
Forty-five lighting companies scattered
throughout the country have purchased
copies mostly in lots of 25, 50 and 100.
Orders have been received from a
number of manufacturers of lamps and
lighting appliances, from contractors and
from several colleges. The primer was
reprinted in full in journals whose com-
bined circulation is about 30,000; ab-
stracts have appeared in more than 150
TRANSACTIONS I. E. S. — PART I
domestic and foreign journals; it was
estimated that altogether about 2,000,000
notices of the primer have been printed
by these journals. The primer was
published in England in a modified
form. A much wider and more general
distribution or circularization is very
probable.
COMMITTEE ON FACTORY LIGHTING
LEGISLATION
The committee, although it had been
appointed recently, drafted and sub-
mitted to the New York Investigating
Commission recommendations on sec-
tions 3 and 4 of proposed Bill No. 18
pertaining to the lighting of factories
and work-rooms. These recommenda-
tions having to do with specifications
for proper and adequate lighting were
made principally in the interests of
ocular hygiene and safety of employees.
The bill in its final form will be printed
in a future issue of the Transactions.
COMMITTEE ON GLARE FROM REELECT-
ING SURFACES
This committee had been in existence
only a short time. In the course of its
work it learned (1) that school-book
publishers are in favor of eliminating
glazed papers; but they require for
books a cheap and durable paper
which will reproduce half-tones well ;
(2) paper manufacturers contend that
they will produce a paper to meet the
requirements of publishers when there
is a demand for it; (3) school officials
have given little or no attention to
glazed surfaces. The committee has
contemplated conducting, in conjunction
with the research committee, an inves-
tigation of the question of glare from
paper; it proposed the starting, at some
future date, of a definite movement to
eliminate polished surfaces wherever
possible, particularly glazed paper from
school books.
COMMITTEE ON RECIPROCAL RELATIONS
WITH OTHER SOCIETIES
Co-operative relations with some
thirty-two professional, scientific, com-
mercial and philanthropic organizations
were promoted during the year. Joint
meetings were held with a number of
others. "It is believed that through the
work of the committee the society's
name and influence have been materially
advanced. The I. E. S. has been
brought to the attention of various
organizations, heretofore ignorant or
indifferent to its existence. While in
every case we have not met with the
success we had hoped, we believe our
work to be cumulative and, if properly
followed up, will ultimately be produc-
tive of most profitable relations with
all organizations interested in illumina-
tion." The report also contained sug-
gestions for the next committee. The
committee promoted co-operative rela-
tions with the following societies :
American Academy of Medicine.
American Association of Cotton Man-
ufacturers (Boston, Mass.).
American Association of Cotton Man-
ufacturers (Charlotte, N. C).
American Electro-Therapeutic Asso-
ciation.
American Gas Institute.
American Institute of Architects.
American Institute of Mining Engi-
neers.
American Ophthalmological Society.
American Medical Association.
American Public Health Association.
The American School Hygiene Asso-
ciation.
American Society of Mechanical
Engineers.
American Association for Conserva-
tion of Vision.
Architects and Engineers Club.-
Architectural Club of Washington.
TRANSACTIONS I. E. S. — PART I
Association of Edison Illuminating
Companies.
Association of Iron and Steel Elec-
trical Engineers.
Association of Railway Electrical
Engineers.
Association of Railway Surgeons.
Association of Stationary Engineers.
Committee on the Prevention of
blindness.
Industrial Safety Association.
Institute of Electrical Engineers.
Medical Society of the State of
Illinois.
Medical Society of the State of New
York.
Medical Society of the State of
Pennsylvania.
Museum of Safety.
National Commercial Gas Association.
National Electric Light Association.
Ohio State Medical Association.
Physiological Society.
COMMITTEE ON PROGRESS
This report was supplementary to the
report of the committee which is printed
in the November issue of the Trans-
actions ; it suggests that it would be
desirable for the next progress com-
mittee to endeavor to base much of its
report on a review of articles pertain-
ing to illuminating engineering which
may appear in the various domestic and
foreign journals; in other words, a
report should constitute as far as pos-
sible a conspectus of progress in illumi-
nating engineering. The preparation of
such a report, it was stated, would in-
volve a vast amount of painstaking
work in abstracting and indexing matter
from many publications.
SECTION DEVELOPMENT COMMITTEE
During the year the committee held
two meetings ; recommended the ap-
pointment of section representatives on
the national papers committee; under-
took the preparation of a guide on sec-
tion management; and suggested the
appointment of representatives or local
secretaries in cities not having sections ;
and in a general way sought to promote
the welfare of the sections and the
society. The appointment of a commit-
tee to organize a section comprising
cities in the Lake Erie region was made
as a result of a recommendation of the
committee.
COMMITTEE ON NOMENCLATURE AND
STANDARDS
The report appears in full in the
December issue (1912) of the Trans-
actions; it gives a number of new
photometric definitions and a brief ac-
count of the committee's activities.
COMMITTEE ON RESEARCH
The report stated that the committee
hoped to be able to submit a report of
accomplishments next year, .if the coun-
cil deems it advisable to continue the
committee. The functions of the com-
mittee were outlined as follows : "It
should be a sort of a clearing-house in
illuminating engineering matters. It
should endeavor to see that investiga-
tions which are necessary for applica-
tion in practise are undertaken by those
competent and prepared to do such
work. It should be prepared to recom-
mend to those desiring to undertake
research work suitable problems to be
investigated. It should bring into closer
co-operation the various scientific and
technical bodies and should keep in close
touch with the various scientific and
technical schools where research work
in any of the allied sciences is under-
taken."
COMMITTEE ON EDITING AND
PUBLICATION
The Transactions for 1912 had ap-
proximately 300 pages less than the
TRANSACTIONS I. E. S. — PART I
Transactions of 191 1, and the cost was
about $1,100 less. Condensation and
elimination accounted for practically all
of both these reductions. Three recom-
mendations were made: (1) the ar-
rangement of the Transactions into
two parts — a news section and a section
devoted entirely to papers, discussion
and reports; (2) the use of a paper
freer from glare than the paper in use;
(3) the publication of a guide setting
forth the requirements and general
style of papers and discussion which
would reduce printing expenses and
facilitate the work of publication.
COMMITTEE ON ADVERTISING
The advertising revenue of 1912 was
$1,356, as compared with $1,225 for 1911.
Contracts obtained for additional space
in the Transactions together with
those pending should net about $2,500
in 1913.
COMMITTEE ON NEW MEMBERSHIP
A quiet but conservative campaign
was made for new members. Most of
the 201 applications and 3 reinstate-
ments during the year 1912 are attrib-
utable to the activities of the commit-
tees. The committee recommended that
the next committee be made up of the
chairmen of the section membership
committees and such others as may be
expedient in the conduct of the com-
mittee's work. If the committee will
keep in close touch with the section
membership committees and encourage
an exchange of ideas or plans, it was
said that its work will be greatly facili-
tated.
1912 CONVENTION COMMITTEE
This report was the last of a series
of reports by the committee. It in-
cluded a scrap book of samples of all
the literature of the recent convention
and a number of valuable suggestions
as to procedure, which should be of use
to future convention committees. A
check for $177.74, the excess of receipts
over expenditures by the committee,
accompanied the report.
Council Notes.
JANUARY COUNCIL MEETING
Twenty-five applicants were elected
to membership at a regular meeting of
the council which was held in the gen-
eral offices of the society, 29 West
Thirty-ninth Street, New York, Jan-
uary 10, 1913. The names of those
new members are listed on page 5.
Twenty-nine resignations, most of
which had been held over from the pre-
vious year, were accepted.
A series of amendments to the by-
laws, most of which were necessitated
by a recent constitutional change in the
fiscal year of the society, was read for
the first time.
Annual reports were received from
the following committees : new mem-
bership, finance, nomenclature and
■ standards, editing and publication, re-
search, reciprocal relations with other
societies, section development, glare
from reflecting surfaces; progress,
illumination primer, factory lighting
legislation. A synopsis of each report
is given elsewhere in this issue.
The final report of the 1912 conven-
tion committee was received.
President Lansingh reported progress
in the work of organization of a Lake
Erie section of the society.
Mr. W. R. Addicks, president of the
American Gas Institute, invited the
council to appoint a representative of
the society to a committee which is to
arrange for a gas congress in San
Francisco during the Exposition in 1915.
TRANSACTIONS I. E. S. — PART I
The president was authorized to appoint
this representative.
The 1912 annual report of the gen-
eral secretary was received. This re-
port with certain modifications was
made the report of the council to the
society; it appears in full elsewhere in
this issue of the Transactions.
Those present at the meeting were :
V. R. Lansingh. president; T. D.
Israel. A. J. Marshall, P. W. Cobb,
George S. Barrows, H. E. Ives, J. \V.
Cowles. R. C. Ware. W. J. Serrill. J. T.
Maxwell. L. B. Marks, C. J. Russell,
E. P. Hyde. C. H. Sharp, E. B. Rosa,
A. E. Kennelly, C. O. Bond. Norman
Macbeth, and Preston S Millar, general
secretary. Mr. W. R. Addicks. presi-
dent of the American Gas Institute, was
present upon invitation.
SPECIAL COUNCIL MEETING
A special meeting of the council, the
first meeting of the new administration.
was held in the Engineers' Club, New
York City, January 10, 1913. Those in
attendance were :
Preston S. Millar, president ; J. D.
Israel, general secretary; C. J. Russell.
George S. Barrows. C. H. Sharp, E. B.
Rosa, W. J. Serrill. C. O. Bond. P. W.
Cobb, R. C. Ware. L. B. Marks. E. P.
Hyde. H. E. Ives. J. T. Maxwell, J. W.
Cowles, Norman Macbeth, and V. R.
Lansingh.
After a discussion of the question of
sustaining membership, it was resolved
that the dues of sustaining members be
left to the discretion of the committee
(committee to be appointed by the
president) having this matter in charge,
that the amount of such dues be not
published in the by-laws, and that the
rules and regulations which may be pro-
posed by the committee be first approved
by the council or its executive com-
mittee.
A list of changes in the by-laws, most
of which were necessitated by the recent
constitutional change in the fiscal year
of the society, was read a second time
and adopted.
President Millar announced his ap-
pointments to various standing and
temporary committees. The appoint-
ments were approved. It was under-
stood that additional appointments to
these committees would be made later.
A list of committees and the personnel
of each appears in the front part of this
issue of the Transactions.
An appropriation of $100 was author-
ized to cover the cost of printing a
prospectus on the society and its work.
The president was authorized to
appoint a committee of five to investi-
gate, and report to the February council
meeting, the matter of appointing rep-
resentatives or local secretaries in cities
not having sections of the society.
It was decided to hold the regular
meetings of the council during the pres-
ent administration in the morning of
the second Friday of each month —
except, of course, during the months of
July, August and September, when no
regular meetings are held.
New Members.
At a meeting of the council held in
New York, January 10, the following
applicants were elected members of the
society :
Arrighi, Roswell.
Agent, The New York Edison Com-
pany, 124 West 42nd Street, New
York.
Baker, Cyrus Rex ford.
Incandescent Lamp Specialist. Gen-
eral Electric Co., 30 Church Street,
New York.
TRANSACTIONS I. E. S. — PART I
BlERMAN, CHAS.
Telephone Engineer, Wisconsin
Telephone Co., 183 gth Street, Mil-
waukee, Wis.
Brown, Melvin P.
Lighting Inspector, Dept. Water
Supply, Gas & Electricity, 13-21
Park Row, New York.
Campbell, O. M.
Sales Engineer, National X-Ray
Reflector Co., 235 W. Jackson
Boulevard, Chicago, 111.
Cole, Chas. M.
Illuminating Engineer, Wheeler Re-
flector Co., 156 Pearl Street, Bos-
ton, Mass.
Conner, George C.
Engineer, National Electric Lamp
Association, 441 1 Hough Avenue,
N. E., Cleveland, Ohio.
Dalton, Parker C.
Salesman, Philadelphia Electric Co.,
1000 Chestnut Street, Philadelphia,
Pa.
Duane, Dr. Alexander.
129 East 37th Street, New York.
Ferguson, Joseph Simpson.
Student, Philadelphia Trades
School, 12th and Locust Streets,
Philadelphia, Pa.
Galavan, Edward.
Sales Engineer, National X-Ray
Reflector Co., 235 W. Jackson
Boulevard, Chicago, 111.
Keane, H. P.
Sales Engineer, National X-Ray
Reflector Co., 235 W. Jackson
Boulevard, Chicago, 111.
Knight, J. Harmer.
Draftsman, Philadelphia Electric
Co., 10th and Chestnut Streets,
Philadelphia, Pa.
La Belle, John N.
Supervising Engineer, National
X-Ray Reflector Co., 235 W. Jack-
son Boulevard, Chicago, 111.
Lewis, Dr. F. Park.
454 Franklin Street, Buffalo, N. Y.
Martin, W. G.
Engineer, National X-Ray Reflector
Co., 235 W. Jackson Boulevard,
Chicago, 111.
Mass, Herbert C.
Illuminating Engineer, 436 Henry
Bldg., Seattle, Wash.
Maxwell, C. M.
Electrical Draughtsman, R. D. Kim-
ball Co., 15 West 38th Street, New
York.
McKinnie, E. C.
Engineer, National X-Ray Reflec-
tor Co., 235 W. Jackson Boulevard,
Chicago, 111.
Parrott, Robert.
Sales Engineer, General Electric
Co., 30 Church Street, New York.
Selleck, John K.
Engineer, National X-Ray Reflec-
tor Co., 235 W. Jackson Boulevard,
Chicago, 111.
Seymour, F. W.
Lighting Inspector, Dept. Water
Supply, Gas & Electricity, 13-21
Park Row, New York.
States, Wilmer M.
Salesman, General Electric Co.,
Edison Lamp Works, Harrison,
N.J.
Sullivan. J. B.
General Electric Company, Foreign
Department, Sarmiento 531, Buenos
Aires, Arg.
Tolman, W. H.
Director, Museum of Safety, 29
West 39th Street, New York.
Section Activities.
CHICAGO SECTION
A regular meeting of the Chicago
section was held in the auditorium of
the Western Society of Engineers,
Chicago, January 15, 1913. Sixty-five
TRANSACTIONS I. E. S. — PART I
members and guests were present. Mr.
T. H. Aldrich of the National X-Ray
Reflector Company and Mr. J. P. Malia,
chief electrician of Armour & Com-
pany, presented a paper entitled "Indi-
rect Illumination of General Offices."
The paper was for the most part a
detailed description of the lighting in-
stallation in the general offices of
Armour & Company.
The following program of meetings
has been arranged :
February 22 — A joint meeting of
engineers, architects and ophthalmolo-
gists, several organizations participating,
in Milwaukee, Wis.
March 19 — Announcement will be
made later.
NEW ENGLAND SECTION
A regular meeting of the New Eng-
land section was held in the auditorium
of the Edison Electric Illuminating
Company, January 21, 1913. Two papers
were read : one on "Commercial Lenses"
by Dr. H. P. Gage of the Corning Glass
Company, Corning, N. Y. ; the other,
"Problems of Lighthouse Service and
How They Met" by Dr. Raymond
Haskell of the United States Lighthouse
Service. Both papers were illustrated
by lantern slides and were very interest-
ing. Preceding the meeting a dinner
was held at "The Georgian," at which
plans for the general welfare of the
section were discussed.
The following program, subject to
change, has been announced :
February 17 — Joint meeting with the
Boston section of the American Insti-
tute of Electrical Engineers. Papers :
"Ornamental Magnetite Arc Lamps" by
C. A. B. Halvorson of the General Elec-
tric Company, West Lynn, Mass. ; "The
Enclosed Flame Arc Lamp" by W. A.
Darrah of the Westinghouse Electric &
Manufacturing Company, East Pitts-
burgh, Pa. ; "Mercury-Vapor Lamps"
by P. H. Thomas, New York.
March 18 — A demonstration of inte-
rior lighting effects, by Preston S.
Millar.
NEW YORK SECTION
The New York section held a joint
meeting with the National Commercial
Gas Association in the United Engineer-
ing Societies' Building, January 9, 1913.
Two papers — "The Lighting of Taft
Hall in the Auditorium Armory" (At-
lanta, Ga.) by Robert F. Pierce of the
Welsbach Company, and "The Lighting
of the Exhibition Hall, Auditorium
Armory" (Atlanta, Ga.), by J. M. Coles,
were presented. About 175 attended
the meeting. Preceding the meeting
there was an informal dinner at Keene's
Chop House.
The program of meetings for the re-
mainder of the season is as follows :
February 7 — A joint meeting with the
Municipal Art Society at the New York
Arts Club.
March 13 — Joint meeting with the
American Society of Mechanical Engi-
neers in the United Engineering So-
cieties Building, 29 West 39th Street,
New York. Mr. Ward Harrison of the
National Electric Lamp Association will
present a paper on "Industrial Light-
ing."
April 8 — This meeting will probably
be held in the United Engineering
Societies Building. Mr. M. Luckiesh of
the National Electric Lamp Association
will present a paper on "Light and Art."
A paper on "Phosphorescence and
Fluorescence" is also scheduled. This
meeting should be an unusually inter-
esting one.
May 8 — A talk on theater lighting by
Mr. Bassett Jones, Jr., at the Clymer
Street Theater, Brooklyn. During the
past year Mr. Jones has conducted a
8
TRANSACTIONS I. E. S. — PART I
great deal of experimental work in
theater illumination particularly in the
production of stage effects. The mem-
bers of the New York chapter of the
American Institute of Architects will be
invited to attend this meeting. Admis-
sion will be by card.
June 8 — It is planned to. have a joint
meeting and outing of all the engineer-
ing societies in New York.
PHILADELPHIA SECTION
A joint meeting with the Philadelphia
section of the American Institute of
Electrical Engineers and the Philadel-
phia Electric Company section of the
National Electric Light Association was
held at the Engineers' Club, 1317 Spruce
St., Jan. 13. Short talks on the subject
of "Modern Illumination" were given by
Prof. George A. Hoadley of Swarth-
more College, Prof. Arthur J. Rowland
of Drexel Institute, and Joseph D.
Israel of the Philadelphia Electric Com-
pany. Dr. G. S. Crampton, W. E. Rob-
ertson, H. Calvert, B. Frank Day, R. F.
Pierce, and G. H. Swanfeld also gave
brief talks on various phases of the
subject of illumination Mr. R. B. Ely
exhibited a number of modern illumi-
nants.
A number of members of the Phila-
delphia section attended the meeting of
the Franklin Institute and the Philadel-
phia Electric Company section of the
National Electric Light Association in
the auditorium of the Institute, 15 South
Seventh Street, Thursday evening, Jan-
uary 30. At that meeting Dr. E. P.
Hyde presented a paper on "The Phys-
ical Laboratory of the National Electric
Lamp Association." Dr. Hyde's paper
was supplemented by a series of lantern
slides showing the new buildings of the
association which occupy a forty-acre
plot in Cleveland, which is to be known
as Nela Park. Dr. Hyde stated that
when the buildings are completed there
will be available excellent facilities for
the conduct of the scientific problems
of the lighting industry.
The following meetings have been
scheduled:
February 21— A demonstration of
interior lighting effects by Preston S.
Millar.
During the week of March 23 a meet-
ing will be held in the New Century
Drawing Rooms. Mr. M. Luckiesh of
the physical laboratory of the National
Electric Lamp Association will present
a paper on "Light and Art." Notices of
this meeting will be issued shortly.
The dates and papers for subsequent
meetings will be announced later.
PITTSBURGH SECTION
Mr. H. W. Shalling read an interest-
ing paper on "Department Store Light-
ing" at a meeting of the Pittsburgh sec-
tion, January 24. The paper and its
attending discussion appears in this
issue of the Transactions. The mem-
bers in attendance were the guests of
McCreery & Company, in whose store
the meeting was held. At the conclu-
sion of the meeting a resolution of
thanks to McCreery & Company was
adopted.
The program of meetings for the rest
of the season is as follows :
February — "Gas Lighting" by S. B.
Stewart.
March— "Moving Picture Lanterns
from the Central Station Point of
View" by J. F. Martin.
April — "Railroad Car Lighting" by
J. L. Minick.
May— "Physiological Aspects of Il-
lumination" by W. E. Reed.
June — Announcement will be ~ made
later.
TRANSACTIONS I. E. S.— PART I
9
Annual Meeting.
Sixty-nine members and guests were
present at the annual meeting which
was held in the Aldine Club, Fifth
Avenue and 23rd Street, New York,
January io, 1913. A dinner preceded the
meeting. Brief addresses on the society
and various phases of its work were
made by Dr. A. E. Kennelly of Harvard
University, Mr. T. C. Martin, secretary
of the National Electric Lamp Associa-
tion, Dr. W. H. Tolman, director of the
American Museum of Safety, Dr. C. H.
Sharp of the Electrical Testing Labora-
tories, Mr. L. B. Marks, first president
of the society, Mr. V. R. Lansingh, the
retiring president; Mr. W. R. Addicks,
president of the American Gas Institute,
and Mr. J. W. Lieb, Jr., vice-president
of the New York Edison Company.
During the meeting it was announced
that the following officers had been
elected at the previous election : Preston
S.Millar, president; vice-presidents, Wm.
J. Serrill, J. W. Cowles, J. R. Cravath,
H. S. Evans; general secretary Joseph
D. Israel; L. B. Marks, treasurer; and
C. O. Bond, P. W. Cobb and W. Cullen
Morris, directors. The announcements
were received with applause. It was
also reported that the constitutional
amendments which had been submitted
at the election had been adopted.
Retiring President Lansingh then in-
troduced President Millar, who deliv-
ered his inaugural address on the
progress and functions of the society.
The address is printed elsewhere in
this issue of the Transactions. A
brief address was also delivered by Mr.
Joseph D. Israel, the newly elected gen-
eral secretary.
The report of the council covering the
work of past year was presented in an
abstract form. The full report, the re-
port of the general secretary, is printed
in this number.
A New By-Law.
The following by-law which outlines
the procedure of section nominating
committees was adopted at a meeting of
the council, January 10, 1913 :
The procedure in nominating and electing
section officers shall be as follows, except when
other procedure shall be authorized by the
Council.
A section nominating committee shall be ap-
pointed by the Section Board of Managers each
year. The appointment shall be reported to the
General Secretary. This committee shall con-
sist of five members of whom at least two shall
be past officers of the Section or members of the
Council. Not later that March 15 of each year,
the General Secretary shall notify the chairman
of the committee that it is the committee's duty
to prepare a nomination ticket containing the
names of those whom they deem best suited for
the section offices to be filled at the ensuing annual
election. The report of the committee shall be
prepared in duplicate, one copy shall be sub-
mitted to the chairman of the section and the
other copy shall be delivered to the General
Secretary not later than April 15. The ticket
thus prepared by the committee on nomination
shall be printed and forwarded to all section
members not later than May 5, in connection
with the ballots for election of general officers.
The election of section officers in other respects
shall be carried out in a manner similar to that
prescribed for the election of general officers,
save that a copy of the report of the Committee
of Tellers on the results of the section election
shall be mailed as soon as prepared, to the chair-
man of the section and to the chairman-elect.
This by-law will standardize and
facilitate the procedure in electing sec-
tion officers. It will also eliminate con-
siderable work and expense in connec-
tion with the elections.
10 TRANSACTIONS I. E. S. — PART I
Annual Report of the Finance Committee for the Fiscal Year 1912.
To the Council of the Illuminating Engineering Society:
In accordance with the provisions of the constitution of the society, the
Finance Committee exercised direct supervision over the financial affairs of the
society.
The committee held a meeting each month except during July, August and
September, examined and approved all bills paid by the society, and presented a
written report at each meeting of the council.
The financial condition of the society as of Dec. 31, 1912, is given in the sub-
joined statement of Messrs. William J. Struss & Co., certified public accountants,
who were employed by authorization of the council to audit the books and accounts
of the society.
The auditor's report shows a deficit of $640.07 for the year and an impairment
of surplus amounting to $1,435.63 since Jan. 1, 1912.
The membership of the society at the close of the year was 1,325. At one
time during the year the membership reached 1,470. Based on an average active
membership of 1,350 for the year, the expenses per member were $7.36. The
income, other than that obtained from membership dues, was derived chiefly from
the proceeds of advertising and miscellaneous sales of the Transactions of the
society.
Early in 1912 the committee reported to the council that the society was likely
to face a considerable deficit at the close of the year, and recommended that imme-
diate steps be taken to secure a larger income from the membership. The council
appointed a special committee on "Financial Policy" and subsequently a committee
on "Revenue," to consider ways and means of placing the society on a sound
financial basis and, as a result of protracted discussion of the subject, the plan
of "Sustaining Membership" (now forming part of the constitution) was finally
evolved. If this plan works out as well as expected, the income derived from
the membership at large will be sufficient not only to defray the ordinary running
expenses of the society, but also to meet increased expenses due to expansion of
the activities of the society.
Respectfully submitted,
A. A. Pope,
A. S. McAllister,
L. B. Marks, Chairman.
TRANSACTIONS I. E. S. — PART I 11
STATEMENT OF THE AUDITORS.
Exhibit "A." — Balance Sheet, December 31, 1912.
ASSETS.
Cash-
On hand and in bank $1,918.23
191 3 — New York Section expenses paid 1912 29.25
Accounts Receivable —
1912 dues $ 12.50
Miscellaneous accounts 510.01
Initiation fees 10.00
1912 advertising 402.38
Total 934.89
Property Accounts —
Furniture and fixtures 630.21
Less depreciation — 15 per cent 94-53
Net 535.68
Badges on hand (29) 84.00
Total :. 619.68
Investments —
Northern Pacific and Great Northern Railway Bonds — $2,000 1,920.00
Total $5,422.05
liabilities.
Accounts payable $ 803.57
Advance dues 2,060.00
Advance fees 5.00
December expenses estimated — Exhibit "B," Schedule No. 1 1,080.00
Advance advertising 19-54
Surplus — Exhibit "A," Schedule No. 1 1,453.94
Total $5,422.05
Exhibit "A," Schedule No. i — Surplus Account, Dec. 31, 1912.
Surplus — January 1, 1912 $2,889.57
Duplicate charge in 191 1 9.98
Back dues 10.00
2.009.55
12 TRANSACTIONS I. E. S. — PART I
1911 New York Section expenses $ 62.10
191 1 General Office 60.30
191 1 Transactions 287.66
191 1 Philadelphia Section expenses 28.85
191 1 Chicago Section expenses. 56.49
Dropped (various) in default of fees 74.00
1911 Election expenses (part) 163.85
191 1 Membership Committee (part) 82.29
Deficit for year 1912 (see Exhibit "B") 640.07
I45S-6I
$1,453-94
Exhibit "B" — Statement of Earnings and Expenses for the Year
Ended December 31, 1912.
earnings.
Members' dues $6,872.19
Advertising 1,356.22
Miscellaneous sales of Transactions 586.78
Initiation fees 375-00
Interest on bonds 80.00
Profit on badges sold 21 .00
Members' certificates 8.00
Total $9,299.19
expenses.
Transactions $2,253.29
December expenses estimated (Exhibit "B" — Schedule No. 1).... 1,080.00
General Office (Exhibit "B" — Schedule No. 2) 4,057.34
New York Section 389.83
Chicago Section 251.17
New England Section 208.43
Pittsburgh Section 148.31
Philadelphia Section 291.32
Committee on Illumination Primer 463.15
1912 Convention Committee 420.14
1912 Election expense 1 14.12
Depreciation — furniture and fixtures 94-53
Papers Committee 44-07
Committee on Reciprocal Relations 4.25
Committee on Nomenclature and Standards i5-7o
Joint meetings with other societies 27.43
Annual Meeting 32-75
Authors' advance copies 5.38
Treasurer's expense 16.44
President's expense 9.25
Exchange on checks 12.36
Total 9,939.26
Excess of Expenses over Earnings $640.07
TRANSACTIONS I. E. S. — PART I 13
Exhibit "B," Schedule No. i — December (1912) Expenses, Estimated.
Primer $680.00
Transactions 250.00
Chicago Section 10.00
Philadelphia Section 20.00
Pittsburgh Section 5.00
New England Section 5.00
General Office (part) 50.00
Miscellaneous 60.00
$1,080.00
Exhibit "B," Schedule No. 2 — Analysis of General Office Account
for the 12 Months Ended Dec. 31, 1912.
Salaries : Assistant Secretary and Stenographer $2,210.68
Rent 666.00
Postage 298.57
Telephone and telegraph 152.98
Printing, stationery, etc 332.37
Miscellaneous 396.74
$4,057-34
TRANSACTIONS
OF THE
Illuminating
Engineering Society
JANUARY, 1913
PART II
Papers, Discussions and Reports
[ JANUARY, 1913 ]
CONTENTS - PART II
Inaugural Address of President P. S. Millar i
1912 Report of the General Secretary 6
Department Store Lighting. By H. W. Shalling 17
Tests for the Efficiency of the Eye Under Different Systems
of Illumination and a Preliminary Study of the Causes of
Discomfort. By C. E. Ferree 49
INAUGURAL ADDRESS OF PRESIDENT MILLAR.*
Fellow Members of the Illuminating Engineering Society:
We who are members feel sure that this society is destined
to fulfill an important purpose. The problem which it must
solve is tremendously difficult and almost unique in its com-
plexity. The goal has been pointed out and described by Past-
President Hyde in these words :
The goal of illuminating engineering will have been attained when as a
result of the concomitant development of its component elements, it will be
possible in every case presented, to design a lighting installation which will
be efficient, effective, artistic ; which will produce an illumination correct
in quantity and quality properly balanced as to high-light and shadow, rest-
ful to the eye and harmonious with the form and color schemes involved ;
which will stand the rigorous test of logical analysis, and will appeal to the
highly developed sense of beauty. The goal of illuminating engineering is
the attainment to the ideal application of perfect knowledge.
At the time when the society was organized we were far from
the goal. Many of us did not know what or where the goal is.
Our lighting practise in general was execrable. To-day we are
well started on our way. While the goal is still so far away
that we cannot afford to interrupt our journey to celebrate
progress, yet we may pause a moment to look backward and
contemplate with gratification the improvement which has taken
place in lighting practise and the advance made in our knowledge
of lighting principles. From such contemplation we may derive
inspiration for the great journey still before us.
The history of the society covers seven years. They have
been busy years. Much has been accomplished though that much
seems little when compared with the great work remaining to be
accomplished.
The first year was devoted to establishing the society, bringing
into its membership those connected with various industries and
professions concerned, and securing papers on some of the many
aspects of the subject of lighting, in order to introduce the society
and start it on its way. Through the Herculean efforts of
President Marks the year closed with more than 800 members,
surely a phenomenal record.
* Delivered at the annual meeting of the Illuminating Engineering Society, New
York, January 10, 1913.
2 TRANSACTIONS I. E. S. — PART II
The second year, Dr. Sharp's administration, was devoted to
the establishment of sections, the extension of the society's
influence throughout various cities and the perfection of its
organization. Many vexing problems of an intra-society character
were met and disposed of. The foundation was laid for the
society's later work.
In the third year, Dr. Bell's administration, the society may
be said to have found itself as an organization. The various
officers attended to their functions and a display of team work
made it evident that this is no one-man enterprise, but a well
organized society with many actively interested members.
During the fourth year, under the administration of President
Gartley, internal affairs were further strengthened and the pur-
pose and work of the society was brought before the gas indus-
try in a way to enlist more support than had ever before been
accorded.
In the fifth year our organization was found to be strong
enough to attempt something beyond the continued development
of its internal activities and the lecture course on illuminating
engineering was conceived by Dr. Hyde and carried through
with notable success. Thus was the scope of illuminating engi-
neering defined, a clearer conception of its various elements
placed before the society and the public, and the ground-work
laid for university educational courses in illumination.
The sixth year witnessed an extension of work to include the
education of the public in lighting fundamentals. Work was
started upon a primer of illumination. At the same time another
step was taken toward the establishment of the profession of
illuminating engineering through Dr. Kennelly's scholarly in-
augural address.
During President Lansingh's administration the illumination
primer has been completed and its dissemination begun. The
society has given additional evidence of realization of respon-
sibility to humanity by the appointment of a Committee on
Glare from Reflecting Surfaces. It has sought to discharge its
civic duty through the activity of a Committee on Factory Light-
ing Legislation which has co-operated with the New York State
Labor Commission in devising constructive but safe and sane
legislation on industrial illumination.
INAUGURAL ADDRESS OF PRESIDENT MILEAR 3
Thus we find that the first four years of the society's work
were devoted to perfecting its organization, improving its inter-
nal mechanism, and developing its function as a forum for dis-
cussion. With the fifth year came activity looking toward the
creation of a profession of illuminating engineering. This was
followed quickly by efforts toward public education. A per-
fectly logical course of action, involving first preparation for
the task, and after at least some material progress has been
made within the society as preparation, the application outside
the society of the knowledge acquired in an effort to benefit the
public and establish the profession.
As I see it therefore we have arrived at a stage of develop-
ment where our functions are defined not alone by declaration
but by actions as well. These functions are three in number.
First, to serve as a forum for the presentation and discussion
of technical questions pertaining to light and illumination,
thereby promoting the advance of knowledge and informing the
membership of knowledge acquired ;
Second, to improve lighting practise through the formulation
and application of principles of good illumination and through
the education of lighting practitioners and the public at large in
matters of illumination ;
Third, ultimately, perhaps, to establish a professional basis for
illuminating engineering.
In regard to the, third function, I believe that we are all in
accord in a desire to promote developments which will contribute
toward the establishment of a professional basis for illumi-
nating engineering; but it does not appear to me that the time
is ripe for determined effort in that direction. I believe that
without neglecting this aspect of our problem we may direct our
greatest energies more profitably toward accomplishment in other
directions. Give education, that great panacea, some further
opportunity to create the demand ; advance the boundaries of
knowledge of illumination ; devote efforts to earning greatest
respect for the society ; and all in good time the need for such
a specialty will be generally recognized and it will be found prac-
ticable to establish the desired professional basis.
As a forum, the society is concerned with the science of
4 TRANSACTIONS I. E. S. PART II
illumination. Knowledge must precede application. It was
inevitable therefore that this part of the work should have feat-
ured the earlier years. Thus the mathematics of illumination,
well grounded by Dr. Sharp in his presidential address in 1907,
has been thoroughly developed and has received a full share of
attention in our deliberations. The measuremnet of light as an
element of fundamental importance has been widely discussed,
though much remains for future development. Papers present-
ing experience of lighting practitioners have naturally predomi-
nated and in the aggregate have added much to our store of
knowledge. These are among the many phases of the science
of illumination which have received treatment commensurate
with their importance. But there are some phases which, con-
trary to the desire and in spite of the effort of the successive
administrations, have not been adequately discussed. Among
these are the principles of architecture and decoration as con-
stituting requirements for lighting design. Our science is defi-
cient in this respect, and there is no more important need to be
met than the supply of knowledge to meet this deficiency.
The art of illumination, in the sense of application of knowl-
edge, has been developed rapidly with the advancement of the
science. Newly designed installations, and new lighting equip-
ments are greatly in advance of those of eight years ago. But
the many installations which transgress principles of hygiene,
esthetics or economy are every-day evidence that relatively little
has been accomplished. And the remedy for this we believe is
education — education of our members and others who have to
do with the installation of lighting equipment ; education of
members of organizations who may become interested in improv-
ing lighting conditions; education of students in our universities
and schools ; education of the public at large. Many of us find
ourselves peculiarly sensitive to untoward effects of bad light-
ing because we have studied illumination and recognize bad con-
ditions to which we formerly were unconscious. Our problem
is to educate the American people to be similarly impressed by
bad lighting. When that is done most of the very bad lighting
will be improved. To succeed, the society must address itself
to the task of furthering this educational work upon which
such an excellent start has been made.
INAUGURAL ADDRESS OF PRESIDENT MIELAR 5
The society this year is committed to a policy of expansion.
We are to seek to extend our influence in a number of ways.
The establishment of local representatives in cities where there
is no section is expected to result in a more general knowledge of
our purpose and work, and to promote the application of prin-
ciples of good lighting. In arousing and maintaining local interest
in lighting matters, this is expected to extend greatly the society's
influence for good.
Sustaining membership provides a means whereby lighting com-
panies, manufacturers and others who recognize the value of the
society's work to them may give a limited amount of support in
application of that work. Of equal importance, it gives the
society an opportunity and an incentive to convince other
companies of the actual or potential value of its work and of
the importance of supporting it.
To sum up, it is my view that the society's greatest needs at
the moment are:
1. Increased knowledge of the architectural and decorative
requirements in illumination design.
2. Extension of educational work in its various forms.
3. Successful application of new constitutional provisions for
expanding the influence of the society.
To make some progress along each line is the purpose of the
administration. If material accomplishment is to result, there
must be a continuance of the active loyal support of officers,
committees and members, which has made possible the develop-
ments of the past seven years. Upon obtaining such support I
feel that we may rely with confidence. Fur the rest, "it is not in
mortals to command success" — but we will do more, we will
deserve it. And paraphrasing the Father of our Country we
may say, "Let us raise a standard to which all who are sin-
cerely interested can repair. The event is in the hands of God."
TRANSACTIONS I. F,. S. PART II
1912 REPORT OF THE GENERAL SECRETARY.
Nineteen hundred and twelve, the seventh year of the society's
history, was notable chiefly for a quiet persistent improvement
in internal affairs and a general extension of the society's work
along educational and co-operative lines. The following brief
review of the year's activities records the salient features.
TECHNICAL AFFAIRS
Transactions. — While there is some difference in the dis-
tribution of papers in the Transactions, yet the total number
(40) is so small as to make it impracticable to observe any
trend in the character of the deliberations. A growing tendency on
the part of the Committee on Papers to refrain from printing
papers of doubtful permanent value has reduced the volume of
the Transactions and increased the number of unrecorded lec-
clvassificatlon of papers in first seven volumes of the
Transactions
Of a nature to No. of
No. of Papers interest particularly Papers
2 1 Architects 42
Decorators 29
Fixture Manufacturers 28
Ophthalmologists 27
Manufacturers of Uluminants- 61
Manufacturers of Lighting
Auxiliaries 6r
Lighting companies 55
Illuminating engineers 222
Scientists S7
37
Subject of Papers
Light
Physics 8
Color 5
General 1
Reflection
coefficients 7
Uluminants
Electric 14
Gas • 20
Miscellaneous 3
Lighting Auxiliaries ■ ■ 10
Illumination 118
Principles 50
Artificial: Interiors. 4S
Exteriors. 16
Natural Outdoors • - 2
Indoors ... 2
Units, Standards and
Calculations 31
Photometry 26
Illuminating Engineer-
ing 16
Miscellaneous 6
Total
26,-
1912 REPORT OF THE GENERAL SECRETARY
tures or papers which have been presented before section
meetings.
As during the previous year, discussions of papers have often
been inadequate. It appears advisable to foster more general
discussions of papers, this being the best safeguard against
inaccuracies which are likely to creep into papers in spite of the
best precautions of authors, and papers and editing committees.
Further emphasis upon advance printing of papers and upon
general discussions, whether written or oral, is therefore in
order.
The addition of Volume VII to the Transactions makes the
aggregate number of papers which have been presented before
the society 265. These are distributed as to character and scope
substantially as indicated in the accompanying table.
CONVENTION.
One of the best evidences of increasing strength of the society
is offered by successive annual conventions, each one of which
makes a more impressive showing than the last. The convention
of 1912 is recorded as being eminently successful. The papers
compared favorably with those of previous conventions in
number and quality; the maintained attendance at the sessions
was extraordinarily large in comparison with the registration,
and evidenced keen interest in the proceedings. The spirit mani-
fested toward the society's enterprise was all that could be desired.
SECTIONS
The condition of the established sections is substantially the
same as at the beginning of the year. The Philadelphia and New
York Sections are relatively strong in interest and attendance at
meetings; the Chicago Section is giving evidence of better main-
tained interest than formerly. The New England Section con-
tinues in need of further development. The new Pittsburgh
Section has been eminently successful during its first year. Its
addition strengthens the Society. Its potentialities for the future
seem good.
A proposal to establish a new section at Cleveland has been
considered during the year with adverse conclusions. Out of it
has grown a proposal to form a so-called Middle West or Lake
Erie Section to comprehend the large cities, including Buffalo on
8 TRANSACTIONS I. E. S. PART II
the east, Detroit on the west and Pittsburgh on the south, meet-
ings to be held from time to time in the various large cities of
that territory. This matter is still under consideration, although
it has made a favorable appeal in Cleveland and Pittsburgh. If
such a section is organized in the near future, its effect upon the
Pittsburgh Section will remain to be determined.
The statistics for section meetings are as follows :
Number Papers
Section of meetings presented
Chicago 8 7
New England 6 5
New York 9 7
Philadelphia 8 9
Pittsburgh 5 7
Totals 36 35
COMMITTEES
One of the Society's greatest elements of strength is its com-
mittees, which with few exceptions are gratifyingly active in the
performance of their several duties. Approximately 75 mem-
bers are devoting time and effort to the work of the general com-
mittees, and others are serving upon section committees. The
Society may well be proud of its committee work. The officers
and members of the Council hold in high appreciation the ser-
vices which are rendered to the society and to the cause through
such contributions.
Committee on Papers. — Provision of papers for the annual
convention, and acceptance or rejection of all papers and dis-
cussions are the chief functions of this committee. The splendid
convention papers program attests the excellence of its work
during 1912.
Committee on Editing and Publication. — The routine work for
which this committee is responsible has been performed during
the year by the Assistant Secretary who under the supervision of
this Committee and of the Committee on Papers acts as editor
of the Transactions. Editorial policies and questions of print-
ing and distributing the Transactions are the chief immediate
responsibility of the committee.
Committee on Section Development. — This committee has
directed its efforts toward the improvement and standardization
igi2 REPORT OF THE GENERAL, SECRETARY 9
of section management methods and the extension of the Society's
influence beyond the territory in which the Society is at present
represented.
Committee on New Membership. — A new method of organiza-
tion, calculated to co-ordinate the general and section new mem-
bership work, was tried by this committee during the year. Due
to the fact that section terms of office expire in June, while the
general society terms expire in January, the work of this com-
mittee has suffered somewhat from changing personnel. How-
ever, a steady, quiet pressure has been exerted, which has brought
into the Society more than 150 men who are thought to be
seriously interested in the work and who should be real assets for
the future.
Committee on Reciprocal Relations with Other Societies. —
This committee has communicated with various organizations
during the year, suggesting co-operation through joint meetings,
exchange of representatives on illumination committees, etc.
The committee has been very effective in attaining its object, and
much of the effort, incomplete at this time, is expected to bear
fruit during the coming year. The lack of a suitable Society
conspectus has made the committee's labors unnecessarily arduous
and has reduced their effectiveness somewhat. The notable
achievements of the year may therefore be taken as an indication
of larger accomplishment in the future when better facilities
shall be available for the committee's work.
Committee on Nomenclature and Standards. — This, the sixth
year of the committee's existence, has been its most active year.
The work of defining photometrical quantities has been continued,
and much of our photometrical and lighting terminology which
was in need of standardization has been reported upon during the
year.
In addition to this work, the committee continued its effort,
undertaken two years ago, to foster international co-operation in
standardization of terminology and units. Its tentative proposal
that an international commission be appointed to deal with such
matters met with some opposition abroad on the ground that the
International Photometric Commission (Zurich Commission)
provides a means for effecting international standardization. A
10 TRANSACTIONS I. E. S. — PART II
plan is being developed for expanding the scope of this Commis-
sion and making it thoroughly representative in every respect.
The committee has decided that if the reorganized commission
fulfills the requirements, no better auspices could be obtained for
international standardization and the most desirable course
would be to await such reorganization and stand prepared to co-
operate to the fullest possible extent in the international work
which would then become possible. Pending the consummation
of this plan, all international endeavor along lines of standardiza-
tion is being held in abeyance.
Committee on Research. — Uncompleted committee organiza-
tion and the lack of problems demanding immediate attention
have resulted in a year of little activity for this committee. Con-
siderable attention has been given to consideration of the scope,
functions and personnel of the committee, and a preliminary step
was taken by the chairman in laying down the principles of
research in a paper at the Niagara Falls convention.
Committee on Progress. — This committee discharged its prin-
cipal function in presenting at the annual convention a report on
progress in the field of illumination.
Committee on Glare from Reflecting Surfaces. — Considerable
work in the way of study of its problem has been under-
taken by the committee, but in the few months elapsing
since its appointment, it has been impracticable to carry much of
the work to a conclusion. The end of the year finds the com-
mittee actively engaged in the. promotion of its work, making
reappointment of the present committee an essential in order
that the effectiveness of its operation may not be interferred with.
Committee on Symbols. — This committee was appointed with a
view to ascertaining whether or not there exists a need which it
would be desirable for the Illuminating Engineering Society to
meet by securing co-operation of various organizations, in an
effort to standardize drafting symbols. While the committee's
final report is not yet available, it appears that the need exists, but
there is considerable doubt if the Illuminating Engineering
Society is the organization which should take the lead in an
endeavor to meet it.
IQI2 REPORT OF THE GENERAL SECRETARY II
Committee on Illumination Primer. — The illumination primer,
completed during the summer, has been distributed very generally
among the lighting fraternity, and those whose vocations are
such as to establish a demand for knowledge of matters of
illumination. Much remains to be accomplished however, in the
way of popular dissemination. The primer has been received
with encomiums all over the country and abroad. It is expected
that its influence will be felt in the years to come both through
awakened popular interest in matters of illumination and through
added interest in the society's work.
Committee on Factory Lighting Legislation. — The work of
this committee represents the society's first step toward the per-
formance of a civic duty. Undoubtedly the New York State
legislation on factory lighting will be the better for the influence
which the committee has brought to bear on behalf of the Society.
Committee on Finance. — The financial condition of the Society
is considered and all payments are approved at monthly meetings
of this committee ; also any financial problems which the Council
may refer to it are discussed and reported upon. Due to the
unsatisfactory condition of the society's finances, this committee
has been called upon to devote more than usual thought and time
to the discharge of its duties during the past year.
Committee on Advertising. — It has been the desire of the
Council to decrease advertising in the Transactions. An im-
pending deficit this year led to a temporary departure from that
policy and the committee responded in a very gratifying manner,
securing advertising contracts which now yield a monthly revenue
about 75 per cent, greater than that obtained at the beginning of
the year.
Council Executive Committee. — This committee conducted the
affairs of the society in the interims between council meetings
especially during the summer period.
Board of Examiners. — This board is depended upon to con-
sider and report on applications from territories other than that
which falls under section jurisdiction, and to report to the
Council upon any other applications which may be referred to it
for consideration.
12 TRANSACTIONS I. E. S. — PART II
Committees on Financial Policy and Revenue. — These com-
mittees have considered the financial problem of the society and
have reported to the Council upon ways and means of meeting
the situation.
Convention Committee.- — The notable success achieved by the
convention was due in no small part to the excellent work of this
committee. Technical, social and financial details were handled
in an expert manner without calling upon the society for financial
assistance other than the costs involved in recording and printing
the proceedings. The committee financed the convention and
turned over to the society a small surplus.
Committee on Tellers. — Appointed to count and report upon
the vote of the membership in election of officers and on con-
stitutional amendments, this committee discharged its duties with
promptness and thoroughness.
Committee on Annual Meeting. — Arrangements for the annual
meeting as well as for the dinner which preceded it were entrusted
entirely to this committee with very pleasing results.
BUSINESS AFFAIRS.
It is difficult to make a brief, general statement of the society's
financial condition because of accounting difficulties which render
unqualified comparison with other years impracticable. A reduc-
tion in surplus from approximately $2,900 in January, 1912 to
approximately $1,450 in January, 191 3 shows a decrease of about
$1,450 in surplus as a result of the year's operation. However,
all December 1912 expenses have been included in the 191 2 state-
ment whereas about $800 of the 191 1 expenses appeared in the
191 2 statement. With this amount deducted from the surplus
decrease, there remains an actual deficit of about $650 for 1912.
It is perhaps not ungratifying to report this deficit. To have
reported an increase in surplus for the year would have been
possible only as the result of a degree of stagnation of society
affairs. Some of the new activities which commended themselves
to the Council were undertaken in the face of an assured deficit
because they were considered- well worth while notwithstanding
the unfavorable financial aspect. The inadequacy of members'
dues to meet the ordinary expenses has been recognized by the
Council and reported to the membership during the past few
10,12 REPORT OF THE GENERAL SECRETARY 13
years. The deliberate sense of the Council has been that since
the society's enterprises were worthy of support, they should be
carried out and increased support should be obtained, instead of
curtailing enterprise to the point where the support already avail-
able would prove adequate.
The year's expenses include approximately $1,000 involved in
the production and dissemination of the illumination primer. If
it shall be found feasible to arrange for the extensive dissemina-
tion of the primer on terms which will make it possible for the
society to reimburse itself for this $1,000 expenditure, the year's
operations may be recorded as showing a profit rather than a loss.
MEMBERSHIP.
There has been some reduction in the number of members dur-
ing the year, the defections being larger than the additions to the
roll. The statistics are as follows :
Members January 1 , 1 9 1 2 1,418
Additions during year 206
Defections during year 289
Membership December 31, 1912 1.335
Seven members have been removed from us by death. Their
names are :
George, Thomas L.
General Superintendent, The United Gas Improvement Co., Broad
and Arch Streets, Philadelphia, Pa.
Manning, Wm. J.
District Manager, Philadelphia Electric Co., Gray's Ferry Road and
Carpenter Street, Philadelphia, Pa.
Mayer, Frederick J.
General Manager, Didier March Co., 30 Church Street, New York,
N. Y.
McGlensey, J. F.
Illuminating Engineer, Union Electric Light & Power Co.. 12th and
Locust Streets. St. Louis, Mo.
Morgan, A. J.
Secretary, National X-Ray Reflector Co.. 235 W. Jackson Boule-
vard, Chicago, 111.
Riblet, A.
Superintendent, 14th Street Station, Consolidated Gas Co., 14th
Street and Avenue C, New York, N. Y.
Spillman, A. J.
Chief of Lamp Dept., Philadelphia Electric Co., 1000 Chestnut
Street, Philadelphia, Pa.
14
TRANSACTIONS I. £. S. — PART II
j co
M
CO
>o
~t
0
t^
^5
t^ VO
^o
ON
m on co >*
w
■* tt
LO
>- Tl-
tN
u +*
Co
e
O
M
■>*
lO
M
O
o o
cr.
-r
If} tO N CO
■*
0 ■*
CN o
lO
CO O
o M
tN
CO
o
CM
rT3
tN
tN
co
ON CO
tN
tN
ON
■* »D N N
r^ CO rD i-i
to
ir> on
O
tN
CO M3
tN
o
m Tt- vo w I tj- I I vo I «' I «o
m rO O tN tN
VO O
to tN r-» M On
<S td
21
w rO O tN
rOCOlOrOfNt^lOl ■*! H I r>.
O
CO
o
M
«
w
w
t-H
o
w
o
fcH
«)
M
p
M to to m
^ONtN I t>.Mt^NMl-ICM« I Is*
1/JV3 IO f) H pi N H
tfl
-23 rt
3 '5 1
3 £
J3 u
t s
<2 S So
«) io n
53 2 a -5
&^ * s
O "J CvO a
M -9 a "3
J3 "£ "^ a
u ft « ■-<
bo
.9
be a
3 3
O fcuO
£ '3
bjO ?
a £
.0
CJ
3
hH
U
I)
Ct
K
O
Tj
a cs
t-I ca tri
y M
5 -
'5c
a a
hi e-5
W .2
d j
co a
. a
M
'C to
c a _
, <3 o y.
•S a «g
ft cd s
•a " «
.— ^3 «
a <-> v,
a .
HH^P
0)
O
en s-
£ ft
1912 REPORT OF THE GENERAL SECRETARY 15
The membership record is involved by the fact that bills for
1913 dues were issued early in December and resulted in a num-
ber of resignations which otherwise would not have been received
until January. Resignations are always to be expected when
bills for dues are issued. In this 1912 record, the defections in-
clude such resignations for the beginning and the end of the
year, while in other years they include such resignations only for
the beginning of the year. Corrected for this irregularity, the
membership is found to have about held its own.
An approximate classification of the members by vocation
appears in the preceding table. Such a classification is difficult
and liable to unavoidable error so that the distribution should be
considered no more than indicative.
GENERAL OFFICE.
The last annual report noted the inadequacy of office facilities
in the United Engineering Societies Building, New York. Dur-
ing the past year an adjoining room has been secured which is
now used as a business office, while the original room serves as
the private office of the Assistant Secretary for editorial and
other work, and for the meeting room of the Council, various
committees, and the New York Section Board of Managers.
FOREIGN RELATIONS.
Abroad the year has been signalized by the announcement of
the formation of a German Illuminating Engineering Society
(Deutsche Beleuchtungstechnische Gesellschaft). Anticipating
that the organization of this Society will be perfected during the
coming year, it may be expected that the three societies located in
respectively the United States, England and Germany, will find
many points of common interest and much that is mutually
beneficial in their activities. With the probability that the
reorganized International Photometric Commission will meet the
requirements of international standardization, there remain for
development methods of co-operation in other matters among
the various societies. The Illuminating Engineering Society
through its Council stands ready to co-operate to the fullest
extent with the English and German societies through their
governing boards.
3
l6 TRANSACTIONS I. E. S. — PART II
GENERAL.
The year's experience has indicated some lack in internal
organization. Primarily this will be met if there can be
established closer co-operation among the sections and between
the sections and the Council. In this latter work the respective
Vice-Presidents can be very influential. Among the committees
there has been evidenced a need for better organization. Most
committees need to hold at least occasional meetings, if the
committee's organization is to be perfected and its plans definitely
formulated. With these two general exceptions the organization
of the society appears to be satisfactory.
The opportunities for effective service loom large for the
future. Provisions for making available additional funds with
which to carry out the work of the society should remove the
only serious handicap under which we have labored in the recent
past. The proven loyalty of large numbers of members gives
assurance of means of accomplishment. All things considered,
there is every reason to look for further growth and extension of
influence in the near future.
Preston S. Millar,
General Secretary.
shaleing: department store lighting 17
DEPARTMENT STORE LIGHTING.*
BY H. W. SHAELING.
Synopsis: — This paper, while it is for the most part confined to the
re-designing of a particular lighting installation, outlines the problem
usually encountered in the lighting of large department stores. Lighting
systems, the relative advantages of direct and indirect lighting, uniformity
of illumination, color of light, avoidance of objectionable shadows, main-
tenance, etc., questions which demand consideration in problems of this
sort are discussed briefly. Generally, the re-designing in this case con-
sisted of the substitution of tungsten lamp units for an existing installa-
tion of electric arc and carbon incandescent lamps. Both the old and
the new installations are described in detail. Photometric data, results
of illumination tests made on several floors, diagrams of the test stations
and photographs of the two installations are also included. The installa-
tion changes described have afforded a lighting system which is not only
more artistic and effective, but one that has reduced the operating costs
of the old system more than fifty per cent.
The choice between different systems and different lighting
units is generally made on the basis of, (1) relative efficiency;
(2) relative attractiveness in appearance. These are the funda-
mental considerations, although there are others which are also
of vital importance.
LIGHTING SYSTEMS.
All lighting systems can in general be classed under one of
three classifications, viz., direct, indirect, and semi-indirect.
In practically all lighting systems some portion of the illumina-
tion is received indirectly. In other words a portion of the
illumination is obtained by light reflected from the ceiling or
walls, or both, before reaching the plane of utilization. In direct
lighting when efficiency is important, it is the aim in general to
make the indirect portion of the illumination small, allowing only
enough light to reach the ceiling and walls to illuminate them
to a low intensity, thereby preventing a gloomy appearance.
Indirect illumination is produced by the light coming from a
very large area — the ceiling and upper portion of walls. This
* A paper read before a meeting of the Pittsburgh section of the Illuminating
Engineering Society, January 24, 1913.
1 8 TRANSACTIONS I. E. S. — PART II
gives what is known as diffuseness of illumination. In such a
system no direct light is received on the plane of utilization, the
light source being concealed in an opaque unit.
What has been said in regard to indirect lighting applies equally
well to semi-indirect lighting with this exception, viz., that the
light source is mounted in a translucent rather than an opaque
unit so that some of the illumination is received directly. When
properly designed, a semi-indirect system possesses all the illu-
mination advantages of totally indirect lighting. In addition, it
is more attractive in appearance and avoids the unpleasant effect
of a brilliant ceiling with no visible source of light. The great
danger in the use of semi-indirect lighting is that the translucent
units used will transmit too much light. When too much light is
transmitted, the efficiency is very rarely any greater than with
totally indirect lighting, and the illumination advantages of the
indirect illumination are greatly reduced.
Experience has shown that the best degree of transmission of
light with semi-indirect lighting units is possible when the bril-
liancy of the light unit is approximately the same as the brilliancy
of the ceiling.
INDIRECT COMPARED WITH DIRECT LIGHTING.
Obtaining a large portion of the illumination indirectly has the
following disadvantages as compared with direct lighting.
(i) Lower efficiency; to produce a given illumination requires
about twice as much light with indirect lighting as with efficient
direct lighting.
(2) More rapid deterioration due to the collection of dirt.
(3) A lower degree of perspective, since sharp shadows are
largely eliminated.
(4) An unduly bright ceiling which often gives an unpleasant
psychological effect, especially when the opaque unit of the in-
direct lighting forms a contrast with the brightly lighted ceiling.
For the reasons cited above, and since in the average depart-
ment store a comparatively large area must be illuminated, it
would seem that in most cases the general illumination can be
obtained more economically, efficiently, and with units which will
be sufficiently attractive in appearance, by a system of direct
illumination.
shalung: department store lighting 19
Moreover, it is generally considered by most authorities that
the advertising value or attractive power of a direct is far greater
than that of an indirect system. It is not meant by this, however,
that exposed light sources should be used for this purpose, but
either totally enclosing units or reflectors which practically con-
ceal the light sources in the case of incandescent electric lights.
It is not the purpose of a department store to install lighting
fixtures for display, except in those spaces as may be devoted
to the sale of such goods. The light units should not attract
attention from the goods displayed; the object should be to
provide proper illumination for the display of goods at a reason-
able expense to the owner; the fixtures, of course, should be
sufficiently artistic in appearance and of a character which will
be in harmony with the architectural surroundings.
IMPORTANT CONSIDERATIONS INVOLVED.
Color Value. — The light units employed should give light which
in color approaches as near as possible that of natural daylight,
so as not to distort the colors of the goods displayed. But there
are certain classes of goods which will be used almost entirely
under artificial light such as is found in the home, theatre, and
similar places; the light under which these goods are sold should
approximate that under which goods will be used, which is in
general the incandescent electric lamp.
Due to its high efficiency, the possibility of a more efficient
utilization of its light by the use of properly designed reflectors,
ease of maintenance, and the range in sizes available, the tungsten
filament lamp has gradually displaced the arc lamp for depart-
ment store lighting. While tungsten lamps do not give the same
color values as are given by daylight, the approximation is close
enough for most practical purposes. For particular cases, special
arc and incandescent units, which give a closer approximation
to daylight values, have been developed and are continually being
improved. At the present time, however, they are being used in
specially prepared spaces and not for general purposes.
Avoidance of High Intrinsic Brilliancy. — The avoidance of
glare from exposed brilliant sources is essential. If enclosing
opal or prismatic glassware be used, this effect is reduced prac-
tically to a minimum. If prismatic or opal reflectors are used
20 TRANSACTIONS I. E. S. — PART II
they should be of a deep bowl shape so as to completely screen
the lamp filament from the eye in all its normal positions. It
is advisable also that the lower portion of the bowl of the lamp
be etched or frosted.
Uniformity of Illumination. — The light units should afford
such distribution of light as will produce a reasonable degree of
uniformity of illumination ; glassware which will effect the most
desirable distribution of the light from the particular lamps, or
for any particular arrangement of outlets or class of lighting
service should be selected with care.
Avoidance of Objectionable Shadows. — It is highly important
to avoid objectionable shadows cast by a customer in standing
before a counter. For this reason careful attention should be
paid to the arrangement of light units and their mounting heights.
Efficiency. — Efficiency is here used in the sense of illuminating
efficiency. The reflector type of glassware is generally consid-
ered more efficient than the enclosing type, and is to be recom-
mended wherever the consideration of efficiency is of primary
importance. Enclosing glassware is to be recommended where
maximum attractiveness in appearance is more important than
efficiency.
Cleaning and Maintenance. — Any efficient lighting system
requires careful attention to cleaning and maintenance. Under
conditions such as exist in a large department store as well as
in any first class lighting installation, the reflectors or enclosing
globes must be periodically inspected and cleaned, or the appear-
ance and the illuminating efficiency of the installation will be seri-
ously impaired. Lighting installations require regular — though
perhaps less frequent — cleaning, just as do the store windows.
It will be shown later how in one instance this detail has been
taken care of in a systematic manner and at a very slight cost.
THE McCREERY & COMPANY (PITTSBURGH) STORE
In the summer of 1912 McCreery & Company decided to re-
design the lighting equipment of their store, which is located at
Wood Street and Sixth Avenue, Pittsburgh, Pa. At that time
their lighting equipment consisted, for the main part, of enclosed
carbon arc lamps with a considerable number of carbon filament
\*
Fig. i. — Former lighting equipment of first floor.
Fig. 2.- -Night view showing lighting units on first floor.
Fig. 3.— Third floor under old lighting system.
pig, 4.— lighting unit used on the second, third, fourth, fifth, sixth, seventh and twelfth floors.
shalling: department store lighting
21
and a few tungsten incandescent lamps. Their purpose in
re-designing their lighting equipment was, first, to improve their
lighting by providing a higher intensity and more uniform illu-
mination; second, to reduce their lighting expenses, and, third,
to provide a modern and attractive installation.
The building is 14 stories high, including the basement and the
attic. The ground floor covers a space 100 ft. x 216 ft. (30.48 m.
x 75.84 m.) and each of the remaining floors 90 ft. x 216 ft.
(27.43 m. x 75.84 m.). The inside area of the first floor is
approximately 20,370 square feet (1892.44 sq. m.), and the
Fig. 5.— Photometric curve of the lighting unit (with a 250-watt clear
tungsten lamp) shown in fig. 4.
remaining floors 17,850 square feet (1658.32 sq. m.). The out-
side of the building is finished in white terra-cotta; the interior,
including the fixtures, is of mahogany.
A careful investigation was made of practically all types of
lamps and reflecting devices with the view of determining their
relative efficiencies and applicability for the various spaces to be
lighted. As a result of the investigation, tungsten lamps with
various accessories, which are described further on in the paper,
were decided upon for the different departments.
All changes were made with practically no changes in wiring;
22 TRANSACTIONS I. E. S. — PART II
all the necessary work such as removing the old and hanging
new fixtures was taken care of by the electrical department of
the store, with no outside assistance whatever. One feature
worthy of particular attention and which might well be followed
by others in making such changes, is the fact that all the changes
on each floor were made in a single night, so that when the sales
force arrived the next morning they found a complete new light-
ing equipment with no trace of the dirt incident to the change.
In this way business was not interrupted.
EQUIPMENT OF VARIOUS FLOORS OF THE BUILDING.
Basement (Shoes, trunks and bags). — The lighting equipment
consisted of 150 and 250- watt bowl-frosted tungsten lamps and
prismatic satin finish reflectors of the extensive type suspended
from ceiling by single pendant drops 15 in. (0.38 m.) in length.
Ample and satisfactory illumination was obtained in this space
and no changes were made.
Basement (Delivery department). — The equipment on this
floor consisted of clusters of carbon filament lamps which were
replaced by 18 single units consisting of a 60- watt clear tungsten
lamp and porcelain enameled shallow dome type steel reflector.
First Floor (Men's furnishings, jewelry, leather goods, station-
ery, dress trimmings, toilet articles, etc.). — The lighting equip-
ment of this floor consisted of 34 2-light fixtures and 16 i-light
fixtures supporting 5-ampere enclosed arc lamps, and 5 4-light
decorative brackets on pillars at the rear of the store equipped
with all-frosted carbon lamps.
In view of the fact that ornamental and expensive fixtures
were already installed, an equipment of lighting accessories in
harmony with them was selected. A reflector-ball type of unit of
special design was decided upon ; it consists of an upper prismatic
reflector resting on a stalactite shaped blown globe, the entire
unit being satin finished. A noteworthy feature of this globe
is the fact that distributions approximating the \Vell-known
extensive, intensive, and focusing distributions may be obtained
by varying the position of lamp within the globe. A position was
»*
Fig. 6.— Night view of lighting units on third floor.
Fig. 7.— Scheme for lighting rug-racks on the fifth floor.
Fig. 8. — Eighth floor with old lighting system.
Fig. 9.— Night view of eighth floor showing new lightim
shalung: department store lighting 23
determined upon which would give practically an intensive dis-
tribution.
Each arc lamp was replaced by a 250-watt clear tungsten lamp.
The decorative units on the pillars were replaced by 25-watt
bowl-frosted tungsten lamps and decorative shades having a
satin finished plate over the bottom, with the exception of the
space occupied by lamp bulb.
Second Floor (Bedding and yard goods). — The equipment on
this floor formerly consisted of 53 5-ampere direct current
enclosed arc lamps with clear inner and opal outer globes. A
fairly high intensity of illumination is required on this floor for
matching dress goods and similar purposes. Each arc lamp was
replaced by a 400- watt clear tungsten lamp placed within a 14 in.
(35.56 cm.) 2-piece diffusing glass bowl. This unit1 (see Fig. 4)
consists of a clear stiletto prism reflector mounted over a satin
finished shallow bowl, smooth on the outside and having stiletto
prisms on the interior surface. These units are mounted in very
attractive fixtures having a verde antique finish. The overall
length of the fixture is 42 in. (1.07 m.) so that the bottom of
unit is approximately 10 ft. 6 in. (3.20 m.) above the floor.
Third Floor (Ladies' suits, waists, and furs). — Fifty-one arc
lamps on this floor were each replaced by a 250-watt clear
tungsten lamp unit and fixture similar to those used on the
second floor (Fig. 4). A number of side bracket units equipped
with 60-watt clear carbon lamps were replaced by 25-watt
tungsten lamps.
Fourth Floor (Millinery, ladies' and infants' wear). — In the
millinery department of this floor 12 arc lamps were each replaced
by a semi-indirect unit equipped with a 150-watt clear tungsten
lamp. These units are of a design to -correspond to the archi-
tecture of the room which is of French design of the period
of Louis XIV. A soft diffused light is obtained, which is supple-
mented to a considerable extent by the illumination received from
the units used for lighting the millinery show cases. Thirty-three
1 A " Holophane-Realite " (shown in Fig. 4.).
24
TRANSACTIONS I. E. S. — PART II
arc lamps on the main portion of this floor were each replaced by
a unit such as used on the third floor (Fig. 4).
Fifth Floor (China, bric-a-brac and rugs). — Fifty arc lamps
were formerly used on this floor. Twenty-five were replaced by
units like those used on the second floor (see Fig. 4) equipped
with 150-watt clear tungsten lamps, and twenty-five by similar
units equipped with 250-watt clear tungsten lamps.
A special form of lighting was installed for the rug rack (see
Fig. 7). Twenty-five outlets were provided on the circumference
Vr^iWf
iillP^
>kw^\vw\\\\V>cW
iMttmmm
PMlilli
wM0m
ill 1 1 1 1 till
1111
||
iUwuT\VQ-«T
i3^^^^^%
ic^ll?i^$&;
llMm?
-^s^N^t:
^^^
Fig. 10. — Photometric curve of eight-inch light-density opal* reflector with
60-watt bowl-frosted tungsten lamp.
of a semi-circle, the radius of which was approximately 2 ft.
(0.61 m.) greater than the maximum swing of the rack arms, the
center of the rack being considered the center of the semi-circle.
Outlets were located on 2 ft. 6 in. (0.77 m.) centers. Wiring was
run in metal moulding and small porcelain receptacles having
a metallic bead for use with shade holder were used. Each outlet
was equipped with one 40-watt clear tungsten lamp and a 30-deg.
angle steel reflector having an aluminumized interior surface.
The units were mounted on the ceiling in a pendant position. In
order that the light units should not present too strong a contrast
* "Veluria" reflector of the Holoplane Works of the General Electric Co.
shalling: department store lighting
25
with the color of the ceiling, the exterior surfaces of reflectors
were also finished with aluminum.
Sixth Floor (Mission furniture, wall paper, draperies, art
goods, and ladies' parlor). — Thirty-five arc lamps used for the
lighting of the main portion of the floor were each replaced by
units similar to those used on the third floor. The ladies' parlor
was equipped with four 5-light fixtures using carbon lamps.
The fixtures were retained and 15 and 40- watt clear tungsten
lamps in 6 in. (15.24 cm.) and 7 in. (17.78 cm.) prismatic balls
were substituted for carbon lamps. In the art rooms which are
Fig. 11.— Photometric curve of 60- watt clear tungsten lamp and concentrating
type reflector used in show windows.
decorated in an oriental manner, five light fixtures which were in
use were retained; 25-watt tungsten lamps were substituted for
60-watt carbon lamps and placed within decorative shades similar
to those used in the brackets on the first floor.
Seventh Floor (Men's and boys' clothing and general offices). —
The general illumination of this floor was taken care of by 45
arc lamps, 36 of which were replaced by 400-watt and 9 by
250-watt clear tungsten lamps in special units like that shown in
Fig. 4. As on the second floor, a high intensity of illumination
is required, on account of the nature of the goods displayed. A
feature which was introduced on these two floors to obtain a
26
TRANSACTIONS I. E. S. PART II
whiter light and allow greater facility in the matching of colors,
was the use of lamps rated three volts below the circuit voltage.
It is interesting to note in this connection the excellent results
obtained, also the fact that even though the lamps are burned
above their rated voltage, a life of 800 hours is obtained.
Offices — 21 ceiling outlets equipped with 100-watt bowl-frosted
tungsten lamps with flat opal shades and 24 2-light desk standards
were replaced by 21 ceiling outlets equipped with 150-watt bowl-
frosted tungsten lamps and intensive type reflectors.
y4 ~ 7y/>/ca/ secf/o/t of '<sAo*y wxitfows ^Aotv/'ny /ocof/o/j of /.tfAf i/w'fa.
• /nef/cofes /ocof/o/r of Co/t/m/ra.
© Ihcftcofso fecaf/o/? of 2-*}f>f i/nite
X~ Intf/ccfes Soy /njvtiicA: ///vm/Mf/aft^ff'ewf//?^ ttrere ro/rer>.
Fig. 12.— Plan of first floor showing location of furniture,
pillars, light units, etc.
Eighth Floor (Furniture and victrola department). — This
floor was formerly lighted by 85 6-light cluster units consisting
of 60-watt all-frosted carbon filament lamps mounted under a
large porcelain shade. In the main portion of this floor 80 units
were replaced by a single 60-watt bowl-frosted tungsten lamp
and an 8 in. (20.32 cm.) opal reflector fastened in the present
fixtures.
The victrola department is now illuminated by five 150-watt
clear tungsten lamps in semi-indirect lighting units.
Ninth Floor (Dining room and kitchen). — In the dining room
71 6-lamp cluster fixtures had been installed; each lamp was a
30-watt clear carbon filament lamp. The fixtures were retained;
shalling: department store lighting
27
15-watt all-frosted tungsten lamps were used to replace 30-watt
carbon lamps. Numerous minor changes were also made on this
floor; 60-watt carbon lamps were replaced by 25-watt tungsten
lamps.
In the kitchen 11 arc lamps were replaced by 250-watt clean
tungsten lamps.
S-
e t©
I
-^e-
^
-s'f
® © ©
370 S.SO SO ■
n&r* © *
stz 1 lsao *t;
i
*2t\L*Jf M
.Vt
£08 *** 40S
I
© ©
twin
//x//'co/es /oca f '/on of 7es/ S/of/ons. Ce///nff /re /$/}/, cfO'-O
9 /mf/ca/ej /oca//on 0/ Co/i/mns. f/e/pn/ of '/amps cboye f/oor, \tff-6
® frtd/co/es /oca/zon of //$/// C/n//s. //e/$h/ of /amps adore 7ej/ f/one, 9'-/Q
Fig. 13.— Section plan of first floor showing location of illumination test stations.
Tenth Floor (Furniture, employees' recreation and lunch
room). — Thirty-two arc lamps were replaced by 150- watt clear
tungsten lamps equipped with extensive type prismatic reflectors.
Seventy-six 60-watt clear carbon lamps were each replaced by a
25-watt tungsten lamp.
Eleventh Floor (Buyers' offices, receiving and stock rooms). —
The equipment formerly consisted of 19 arc lamps and 13 4-light
28
TRANSACTIONS I. E. S. — PART II
clusters of 6o-watt clear carbon filament lamps. Thirteen of the
arc lamps were replaced by 250-watt and 6 by 150-watt tungsten
lamps. Each 4-light cluster was replaced by a single 60-watt
tungsten lamp and extensive type prismatic reflector.
Twelfth Floor (Fitting rooms and wash-rooms). — Forty arc
lamps were formerly used. Ten were replaced by 250-watt
tungsten lamps in the special unit shown in Fig. 4 and 30 by
150-watt tungsten lamps and opal shades.
Thirteenth Floor (Attic). — Six arc lamps and 100 60-watt
carbon drop lights were formerly used; 4 of the arc lamps were
replaced by 150-watt tungsten lamps, and 2 by 100-watt tungsten
# /nd/cates /ocof/on of Co/vmns.
© Jnd/'cates /ocof/on of £./gt>f Umfs.
X " Jhdicofes-bqy in tv/i/c/r Ifft/minof/on ffeod/nga ttvse fo/zen.
Fig. 14.— Plan typical of the general arrangement of furniture, light units, etc.,
on the upper floors, with exception of eighth.
lamps. Each 60-watt carbon lamp was replaced by a 25-watt
tungsten lamp.
Show Window Illumination. — An average section of these win-
dows is 23 ft. 6 in. (7.16 m.) long and 8 ft. (5.49 m.) deep, the
ceiling is 19 ft. (5.79 m.) above the bottom of the show case.
The plate glass extends to a height of 12 ft. (3.66 m.). Prism
glass being used at the top. The sides and back of the windows
up to a height of six feet are covered by a dark green velvet
curtain or drapery ; clear glass plates are used above this to let as
much daylight into the store as possible.
The former equipment for windows consisted of a total of
326 60-watt clear carbon lamps mounted in a mirrored trough
placed back of the transom bar. The present equipment consists
shalling: department store LIGHTING
29
of a total of 195 60- watt clear tungsten lamps; 15 are used in each
section, the outlets being located back of transom bar on approxi-
mately 14 in. (0.35 m.) centers. Each lamp is equipped with a
100-watt size concentrating type prismatic reflector ; the units are
mounted at an angle so that tip of the lamp is pointing at a line on
e
© % h % %
"© <5>© '© © ~@
O B ■ B » P
® © @ @ © <k
© % % % ^4"-^
b *® @ ©^ ©
© © © @ ® ©
^@> ^ ^ % %> ^
• Ind/cates /ocafo/i of Cc/</m"s.
$ " //ycf/cafes /ocaf/0/7 of £J$>S?/ Cfo/fe.
■ facf/cofes /ocof/on of Sfo/'/ons oftrA/cA ///i/mjnof/o/7 f?eaa'//7ffS were fafre/i
Fig. 15. — Section plan of third floor showing location of illumination test stations.
floor approximately two-thirds of the distance from plate glass
to opposite enclosing wall of window. This angle is found to
be the best for the average form of window dress, but in excep-
tional cases, provision is made for varying the angle. The light
units are concealed from the view of persons standing on side-
walk by means of an ornamental drapery placed between the
light units and plate glass. The units are wired so that alternate
30 TRANSACTIONS I. E. S. — PART II
lamps are on the same circuit, allowing one half of the units to be
used on ordinary occasions and all units when desired. By using
a larger size reflector ioo-watt lamps may be used on special
occasions when a higher intensity of illumination is required in
any one section for display of special goods. With a wattage
allowance of but 4.7 per square foot (0.09 sq. m.), very satis-
factory illumination has been obtained. It will be noted from
fig. 10 that a very small amount of light is received on the side-
walk; practically all the light is directed on the goods displayed.
In order to keep the light from the upper part of window and
also from interior of the store, a metal strip has been placed above
the light units and painted white on the under side. A small
valance has been suspended from this strip which serves to cut
off the light from the interior of the store. The fact that some
light is received directly on the sidewalk as well as additional
light caused by direct reflection from the goods on display, is
worthy of attention, since such light serves to attract the atten-
tion of the passersby.
Show Case Illumination. — The illumination of show cases in
a large department store is of very great importance and a
large factor in the use of energy. In the present instance the show
cases were formerly lighted by carbon filament tubular type
lamps, but are now lighted far more satisfactorily by the new
tungsten 25-watt drawn wire tubular type lamp which is par-
ticularly suited for this class of lighting.
NOTES ON THE ILLUMINATION TESTS.
After the installation was made, illumination tests were con-
ducted in various departments to determine the intensity and
uniformity of illumination obtained.
Test No. 1 (First floor; buff ceilings — soiled condition; 6 years
since painted — buff walls; dark wood floor; mahogany fix-
tures).— The arrangement of the furniture made it necessary to
select test stations as shown in Fig. 13. Horizontal illumination
readings were taken on a plane 32 inches (0.81 m.) above the
floor, which is the height of the counters throughout store. The
values obtained at various stations are shown in Fig. 13. All
values were corrected for any changes from rated voltage of
lamps which were noted.
shalling: department store lighting
31
Test No. 2 (Third floor; buff ceilings — soiled; walls buff; dark
green carpet; mahogany fixtures). — Test stations were selected
as indicated in Fig. 15, because of few obstructions in this bay.
Horizontal illumination readings were taken on a plane 32 in.
(0.81 m.) above the floor or at the counter level. The following
table indicates readings obtained at various stations :
tation
Foot-candles
Station
Foot-candles
Station
Foot-candles
I
1-93*
15
3-94
29
2.6o
2
2.09
16
4.42
30
2-93
3
2.12
17
3-95
31
2.76
4
2.1 1
18
3.62
32
2.61
5
2.28
19
4.08
33
2.76
6
2.30
20
4.40
34
3-oS
7
1-93*
21
3-94
35
2.60
8
2.84
22
3-63
36
1.89*
9
3-11
23
4.48
37
1.99
10
2-95
24
4-°3
38
2.01
11
2.77
25
3.62
39
2.09
12
2-93
26
4-13
40
2.13
13
3.01
27
4.12
4i
2.28
14
2.84
28
3-63
42
1.89*
* Indicates light partially obstructed by column.
Test No. 3. — The tests were conducted on another floor on
which the change from arc to tungsten lamps had not yet been
made. Tests were made in a bay occupying the same relative
position as bay in test No. 2, and readings were made at the
same stations under practically similar conditions. The readings
obtained were as follows :
Station
Foot-candles
Station
Foot-candles
Station
Foot-candles
I
I.56*
15
2-33
29
I.98
2
I.62
16
2.80
30
2.07
3
1-55
17
2-93
31
i-95
4
I.5I
18
2.72
32
1-77
5
1-45
19
2.67
33
2.32
6
1.49
20
2.38
34
2.17
7
1-33*
21
2.28
35
1.58
8
1.92
22
2.36
36
1.56*
9
2.04
23
3.18
37
1.56
10
2.00
24
2.84
38
1.48
11
1.92
25
2.09 .
39
i-59
12
2.24
26
2-79
40
1.65
13
2.02
27
3-23
4i
1.58
14
i-95
2S
2.52
42
1.39*
* Indicates light partially obstructed by column.
Height of lamps above floor 10 ft. 6 in. (3.20 m. ).
Height of lamps above test plane, 7 ft. 10 in. (2.39 ni.).
4
32
TRANSACTIONS I. E. S. — PART II
Test No. 4 (Eighth floor; buff ceilings — soiled; dark green
walls; dark floor; mahogany fixtures). — Horizontal illumination
readings were taken on this floor at a plane 32 in. (0.81 m.)
above the floor at similar stations and under similar conditions
as Test No. 2, with the exception that four light units are
provided per bay, each consisting of a 60-watt bowl-frosted
tungsten lamps in 8 in. light opal reflectors.1 The results of test
were as follows :
Station
Foot-candles
Station
Foot-candles
Station
Foot-candles
I
0.16*
15
1. 21
29
I.20
2
1. 21
16
1. 21
30
I.32
3
I-I5
17
i-i3
31
I.28
4
1. 17
18
1. 00
32
1. 19
5
1.15
19
1. 11
33
1. 21
6
1. 16
20
1. 16
34
1. 18
7
0.82*
21
1. 12
35
I- 13
8
1. 15
22
1. 19
36
O.86
9
1. 18
23
1.19
37
1.33
10
1.16
24
1.19
38
i-3t
11
1. 12
25
1. 11
39
1.26
12
1. 16
26
1.13
40
1.27
13
1. 17
27
1. 16
4i
1.20
14
1.09
28
1,15
42
1.03*
* Indicates light partly obstructed by columns.
Height of lamps above floor approximately 10 ft. 6 in. (3.20 m.).
Height of lamps above test plane approximately 7 ft. 10 in. (2.39 m. ).
In all the foregoing tests it was necessary to choose bays for
test purposes which were practically free from obstruction, and
while the readings obtained for any one bay if properly con-
sidered will give a fair idea of the average intensity for that
bay, these figures have not been given inasmuch as they might
be construed as representing average intensity for whole floor
space.
CLEANING AND MAINTENANCE OF PRESENT INSTALLATION.
A careful system for regular inspection and cleaning has been
initiated in this store whereby all units in the building are
brushed off with a stiff brush and cloth once every1 two weeks,
and each unit is taken down and washed thoroughly once each
month. The cost of cleaning, including labor and materials, is
approximately $350 per year for approximately 1,000 units
1 Holophane " Veluria."
shalling: department store lighting 33
installed. Since the units are thoroughly cleaned once per
month, the cost of cleaning per unit will be approximately
3 cents, which includes one dusting and one washing. A fair
division of this cost would be y2 cent per unit for dusting and
2]/2 cents for thorough cleaning. Under such a system the
illuminating efficiency is kept at a practical maximum. It is
probable, however, that in other places where dirt and dust con-
ditions are not as severe, that the cleaning could be done at less
frequent intervals. The above-given figures represent a very
small part of the operating expense, and are considerably less
than cost of trimming and other maintenance charges incident
to the operation of the old system.
CONCLUSION.
With the aforementioned changes, this store now has a thor-
oughly modern and attractive lighting equipment, a considerably
higher intensity of illumination at a saving in operating expense
of over 50 per cent. The figures given for cleaning and main-
tenance serve to indicate what may be accomplished by syste-
matic attention to this detail at a cost probably not as great as
the cost of cleaning the windows.
The writer desires to take this opportunity of heartily thank-
ing Messrs. L. J. Kiefer, of the McCreery Company, and P. C.
Keller for the valuable assistance which they rendered him in
collecting the data for this paper.
DISCUSSION.
Mr. G. H. Stickney : This paper contributes some important
data on the recent practise in department store or rather, as I
should class McCreery's Store, dry goods store lighting. It is,
therefore, of great value to all those who have lighting problems
of this character.
A large dry goods or department store combines in one institu-
tion a large number of departments or stores. As such, each
department has its own lighting requirements pertaining to the
class of merchandise handled. Thus, we find the problems of
the silk store, furniture store, picture store, jewelry store,
restaurant, etc., but in addition, since these are all brought
together in one establishment, the lighting of each must bear a
34 TRANSACTIONS I. E. S. — PART II
relation to all the others in order that we may have a unity and
dignity in keeping with the large institution.
Referring to some detail points in the paper, I note that the
author believes that the brightness of the lighting units should be
equal to that of the ceiling. From my own observation, I am led
to the opinion that, for appearance sake, it is desirable in an in-
stallation of this type to have the units preferably brighter than
the ceiling, although this should not, of course, be carried far
enough to introduce objectionable glare.
As to color values, there has, in the past, been considerable
difference of opinion regarding the requirements of dry goods
stores and the practise in different cities has differed. There are
certain departments, such as dress goods and silk departments,
where accurate color selection is desirable; both daylight and the
prevailing evening light should be available so that goods for
street wear may be selected by daylight and those for evening
wear by artificial light. One condition which is often overlooked
is that a high intensity, say 20 foot-candles or more, is necessary
for accurate daylight selection of delicately colored materials.
Moreover, the only accurate color matching artificial lights are
all relatively inefficient and, as a result, it is not practicable, in
the present state of the art, to light an entire department by
artificial light for daylight color matching. The best solution of
this problem is that adopted by the McCreery Company, namely,
that of providing small rooms or booths where artificial daylight
of the proper diffusion and intensity can be obtained in the even-
ing, and correspondingly, evening light in the daytime. Even
where color matching booths are not provided, nearly all the large
dry goods and department stores are turning to the tungsten fila-
ment lamp. Among these may be mentioned Marshall Field
Co., Carson Piru Scott & Co., Mandel Brothers, Rothchilds and
The Fair in Chicago, and the McCreery Co., Gimbels and Green-
hut-Siegel-Cooper Co., in New York. There are many others
nearly, if not equally prominent.
As expressed to me by the manager of one of the large dry
goods stores in Chicago, the pleasing appearance of his store
under the tungsten filament light much more than compensated
for the color matching advantages of any other light. There are
shaujng: department store lighting 35
relatively few departments in such a store where accurate color
matching is necessary, while in other departments, the warm
color tone is considered so important that it is often the practise
to exclude daylight by mean of shades or screens and display the
goods throughout the day by the warm light from the tungsten
tilament lamp.
The enterprise of this company in making the new installation
without in any way interfering with the sales activity of the store
is commendable. I know that this meant considerable planning
beforehand. In many stores even the simple maintenance is not
planned so as to avoid such interference.
Another point worthy of mention is the excellent clean con-
dition in which we found the installation in our inspection to-night.
YVe cannot emphasize too much the importance of cleanliness in
maintaining the efficiency and appearance of any lighting system.
The practise of regular inspection and cleaning at prescribed
intervals is the only way to insure this result, and, as brought out
in the paper, the cost, even in Pittsburgh, is surprisingly low.
The whole installation is admirably adapted for its purpose and
is one of the best I have seen anywhere.
Mr. W. M. Skiee : With reference to showcase lighting, I
would like to call attention to the reduction in heat in showcases
due to the use of tungsten filament lamps. With a 50 watt
tubular gem filament lamp the glass on showcases is often heated
and the breakage of the glass, due to expansion, is common.
With the tungsten filament lamp of half the wattage 28 per cent,
more candle-power is obtained than with the carbon lamp, and
there is less liability of breakage of the glass showcase due to
heat.
In connection with the color of light, my attention was recently
called to a theatrical company which made a practise of carrying
an equipment of strip and border lights- composed of 32 candle-
power carbon filament incandescent lamps, part of which were
amber dipped. This was departing to a considerable extent from
daylight values and the reason given was that the make-ups were
artificial and, therefore, required other than daylight values to
produce the best results. It is no doubt true that daylight value
36 TRANSACTIONS I. E. S. — PART II
in stores is not necessary for the proper illumination of materials
which will be used in the home or elsewhere under incandescent
electric lamps.
Mr. S. G. Hibben : With reference to the question of main-
tenance, it would seem that in retail sales houses where the clean-
ing is not as thorough or does not follow as definite a schedule as
in this store, that of reflectors of equal efficiency the one which
will show dirt, and consequently the need of proper cleaning and
up-keep, is preferable.
On going through the store I also noticed considerable glare
from the upper parts of the prismatic units, and it is suggested
that frosting of these upper portions would be highly advisable.
A commendable point I wish to call attention to is the fact that
the new units on most of the floors are placed higher than were
the replaced arc lamps. In the low position, the arc lamps were
decidedly unpleasant.
On the page referring to window lighting it is stated that a
large reflector is used so that small or large lamps may be em-
ployed as desired. I would like to ask if it is not true that in this
case, where the small lamp is used in the reflector designed for
the next larger sized lamp, that the candle-power distribution
is not changed, since the filament position is different than that
for which the reflector was originally designed.
I note also that on different floors the same sized glass
accessories are used, but different sized lamps are employed. I
believe the author has neglected to mention, in this connection,
the very desirable feature of these fixtures is that there is pro-
vision made for a mechanical adjustment to allow of nearly the
same distribution when using various sizes of lamps.
It must be remembered that the paper cannot be used as a basis
of design, since it is essentially a description of an installation.
Some of the main units employed are admittedly not the most
efficient that might have been obtained. They were not selected
on account of color values-; their maintenance cost is not low; nor
is their deterioration small. It is said that their harmonizing and
decorative appearance determined their selection, and of course
when one discusses taste there is a wide latitude of opinion.
shaixing: department store lighting 37
Mr, Ward Harrison : With reference to maintenance, we
have found that with open reflector equipment approximately
one-half of the deterioration in the illuminating value of the
reflectors arises from the deposit of dust on the outside of the
reflectors, and one-half is due to dust on the inside and on the
lamp bulb. When the lamp is enclosed as in the McCreery fix-
tures, one-half of this deterioration is largely eliminated. This,
with the fact that a clear instead of a bowl-frosted lamp is used,
makes up in a large measure for the lower efficiency to be
expected because of enclosing the lamp.
The complete change of the lighting equipment on a floor in
one night is a commendable plan because of the fact that the
contrast in the colors of the arc and incandescent lamps is not
pleasant and often store employees are not in favor of the change
when they see the two systems thus compared. Both illuminants
vary from daylight ; one has an excess of red, the other of blue
rays ; and in juxtaposition the effect is not pleasing.
In passing through the store I noticed that the ceilings were
of such a color that their coefficient of reflection was not high.
Had they been painted white, the efficiency of the installation
would have been increased appreciably and the contrast in bright-
ness between light sources and the ceiling would also have become
less marked.
With reference to the illumination tests. I would like to ask
Mr. Shalling whether the foot-candle values given are in each
case representative of the whole floor area, and also at what
period during the life of the lamps the tests were conducted; in
a word, whether the data as given are representative of average
conditions including depreciation due to dust.
Mr. L. J. Kiefer : The old equipment consisted of standard
enclosed carbon arc lamps, not intensified arc lamps.
The heat formerly given off by carbon lamps in showcases
where jewelry was displayed was such that injury frequently
resulted to jewel settings, so much so that it was a common thing
to send material back to the factory to have the stones reset. The
use of high efficiency tungsten lamps has eliminated this trouble.
The question of color values has been carefully considered in
connection with the lighting of the McCreery & Company store.
38 TRANSACTIONS I. E. S. — PART II
On the second floor there is provided convenient space next to
the windows where clerks may show silks and dress goods under
daylight. For dark days and during a few hours in the evening
in some seasons, there is also a special room equipped so that a
customer may see goods under both evening and daylight con-
ditions. An intensified arc with a color screen illuminates the
room, forming a circular panel of white light on the ceiling.
Incandescent lighting is also provided and can be conveniently
switched on as a customer desires.
Mr. H. W. Shaixing (in reply) : The units formerly used in
this store were not the intensified type of arc lamps, but were
standard enclosed carbon arc lamps.
In designing the installation, it was found by careful experi-
ments that satisfactory color matching for most practical pur-
poses could be obtained by the use of the tungsten filament lamp.
In some cases a much closer approximation to true daylight
values was obtained by burning the lamps above their rated
voltage. For particular cases, daylight can be used as a general
rule by carrying goods to the windows ; a booth is also available
with a specially prepared color matching device.
In answer to Mr. Harrison's questions, I would say that the
areas chosen for test purposes were representative; all lamps on
any one floor were burning during the tests. Slightly different
intensities might be obtained near the windows or side walls, but
an average of the values found can be considered to be practically
an average for the entire floor space. The units had been in use
about three months when tests were made.
In the cut glass department, clear reflectors and clear lamps are
used. In the showcases a metal reflector having a polished sur-
face is used. In the millinery showcases, polished metal
reflectors are also used.
In answer to Mr. Hibben's remarks, it will be noted that in the
clothing department where 400 watt units are employed with
lamps burned above their rated voltage, that excellent color
values are obtained while the glare is not at all objectionable or
even noticeable to the average customer, considering the time
such customer would usually spend in this department.
On the first floor in order to obtain the required intensity and
shaeung: department store lighting 39
to get the desired ornamental effects, it was necessary to use
about twice the wattage ordinarily used on the other floors.
In designing an installation for a store of this character it is
necessary to carefully analyze conditions on the various floors in
order that there will not be too great a sameness to the units
employed. The units should be in harmony with the architectural
and fixture conditions.
40 TRANSACTIONS I. E. S. PART II
TESTS FOR THE EFFICIENCY OF THE EYE UNDER
DIFFERENT SYSTEMS OF ILLUMINATION AND
A PRELIMINARY STUDY OF THE CAUSES
OF DISCOMFORT.*
BY C. E. FERREE.
Synopsis: — Besides outlining (I) the problem which confronts the
investigator who would determine the effects of various lighting systems
on the eye, this paper discusses: (II) the scale or general level of effi-
ciency of the eye under different systems of lighting, with brief comments
on the conventional tests for the efficiency of the eye such as, (a) color
discrimination, (&) brightness discrimination, (c) visual acuity — the latter
tests, modified, it is contended are adequate for the determination of the
general level of efficiency of the unfatigued eye; (III) loss of visual
efficiency as the result of a period of work — here it is contended that
each of the aforementioned tests fails to show a true loss of visual effi-
ciency, and a new test is described. The paper is concluded (IV) with a
brief statement of some of the causes of ocular discomfort under various
conditions, and a description of a method of making a comparative esti-
mate of discomfort.
I. INTRODUCTION.
In 191 1 the American Medical Association appointed a com-
mittee to study the effect of different lighting systems on the eye.
The writer was asked to share in the work of this committee.
The problem presented to him was to furnish tests that would
show the effect of different lighting systems on the eye and more
especially to devise, if possible, a test that would show a loss of
efficiency as the result of three or four hours of work under an
unfavorable lighting system. It is the purpose of the following
paper to give a preliminary report of the work that has been
carried on by the writer in this field during the past year.
Confronting the problem of the effect of lighting systems on
the eye, it is obvious that the first step toward systematic work
is to obtain some means of making a definite estimate of this
effect. The prominent effects of bad systems of lighting are loss
of efficiency, temporary and progressive, and eye discomfort.
Having devised methods which after six months of testing he
has found to be accurate and practicable, the writer has under-
* A paper read at the sixth annual convention of the Illuminating Engineering
Society, Niagara Falls, Ont., September 16-19, J912-
ferree: tests for the efficiency of the eve 41
taken to determine (1) the lighting conditions that give in gen-
eral the highest level or scale of visual efficiency, (2) the condi-
tions that give the least loss of efficiency for continued work,
and (3) the conditions that cause the least discomfort. This
plan of work, it is scarcely needful to remark, will involve a
wide range of experimentation. The crux of the problem, as the
writer conceives it, is, however, to secure reliable methods of
estimating effect. Having these methods, the factors whatever
they may be, intensity, quality, position of light relative to the
eye, etc., can be varied one at a time and the effects be determined.
From these effects it should not be difficult to ascertain what
lighting conditions are best for the eye and what is the relative
importance of the factors that go to make up these conditions.
Further it should be possible on the practical side to test out and
perfect a lighting system, so far as its effect on the eye is con-
cerned, before we put it on the market.*
In this report nothing more will be attempted than to indicate
what methods may be used in the three steps of the problem as
outlined above.
II. THE SCALE OR GENERAL LEVEL OF EFFICIENCY OF
THE EYE UNDER DIFFERENT SYSTEMS OF LIGHTING.
A general survey of the field shows that at different times the
following tests have been used in one capacity or another for de-
termining the efficiency of the eye : brightness discrimination, col-
or discrimination, and visual acuity. No extensive use, if any at
all, has been made of any of these with the exception of visual
acuity in connection with problems of the type here considered,
but the fitness of their application in some form to such problems
is evident at a glance. If the eye's efficiency is to change at differ-
ent times and under different conditions of lighting, it should be
manifested in changes in brightness discrimination, color discrimi-
nation, or visual acuity. The first step in our work would, then,
seem to be to devise for these points tests which are sufficiently
sensitive for use in work of the kind we have in hand. The gen-
eral nature of these tests is too familiar to need detailed mention
here. A few special points may. however, be given in passing.
(1) The threshold or limen test is the most sensitive and practical
* This latter point was suggested to the writer by reading Dr. Ives' discussion of this
paper (p. 57).
42 TRANSACTIONS I. E. S. — PART II
for color sensitivity. In making this test the pre-exposure1 and
the surrounding field2 should be of a gray of the brightness of
the color at or near its threshold value. Further, the illumination
of the room must be kept constant from test to test.3 If the
colored light is to be obtained by reflection, disks of standard
1 By pre-exposure is meant what the eye rests on immediately preceding its stimula-
tion by color. It is obvious that there must always be some pre-exposure and, unless
care be taken to eliminate its effect, it will influence the eye's sensitivity to color. Even
closing the eye, as is often done before stimulating by color, is the equivalent of giving a
black pre-exposure. All color must of course be eliminated from the pre-exposure. It
should also be of the same brightness as the color by which the e3re is to be stimulated. If
not it gives a brightness after-image which mixes with the succeeding color impression
and reduces its saturation. This reduction of saturation takes place apparently at some
physiological level posterior to the seat of the positive, negative, and contrast color
processes commonly supposed to be located in the retina. (See Ferree and Rand : " The
Fusion of Brightness with Color — The I,ocus of the Action," Journal of Philosophy, Psy-
chology and Scientific Methods, VIII, 1911^.294.) If the pre-exposure is lighter than the
coior it adds \>y after-image a certain amount of black to the succeeding color impression
and, if darker, it adds a certain amount of white. Since white inhibits color more than
black, the effect of a dark pre-exposure is to reduce the sensitivity to color more than the
effect of a light pre-exposure. But since both white and black as after-effect reduce the
sensitivity to color, the eye is rendered more sensitive when no after-image is given, i. e.
when the pre-exposure is of the same brightness as the color. The pre-exposure therefore
should be to a gray of the brightness of the color. No brightness after-image will be
added thereby to the succeeding color impression to modify either its saturation or color
tone.
2 When the surrounding field is either lighter or darker than the color, brightness is
induced by contrast across the colored surface. When the surrounding field is lighter
than the color, a certain amount of black is induced, and when darker, a certain amount
of white is induced. As stated above, the mixture of this white or black with color,
although it does not alter the amount of colored light coming to the eye. reduces the
saturation of the color. The effect of brightness contrast can be eliminated only by
making the brightness of the surrounding field a gray of the brightness of the color.
This can be done by means of a gray screen around the color, or by a larger gray disk in
case a color mixer is used.
3 In case the colored light used for the stimulus is obtained by reflection from a
pigment surface, a change in the general illumination of the field of vision affects the
results of the sensitivity tests in the following ways. (1) It changes the amount of
colored light coming to the eye. (2) By changing its brightness adaptation it changes
the sensitivity of that part of the retina upon which the colored light falls. (3) By
changing the sensitivity of the eye to brightness after-image and contrast, it changes the
amount of brightness added to the color as the result of pre-exposure and surrounding
field, and therefore changes the effect of pre-exposure and surrounding field upon the color
impression. Moreover, the effect of pre-exposure and surrounding field cannot be elimin-
ated even when both are made of the brightness of the color for some given illumination,
unless that illumination be kept constant throughout the test for, when it changes, the
brightness of the color and of the grays used as pre-exposure and surrounding field does
not change in equal amounts ; hence, the brightness equality which is needed cannot be
maintained. In case the colored light is not gotten by reflection from a pigment surface
but is obtained from monochromatic sources from standard filters or from the spectrum,
only the last two of the factors stated above influence the results of the tests for color
sensitivity. In the tests made by the writer, the general illumination was rendered
constant by methods to be described later in the paper.
Although for the purposes of this work the tests for color sensitivity could never be
conducted in the dark-roam, still it may be of general interest to note at this point that the
elimination of the effect ot pre-exposure and surrounding field cannot be accomplished
in work on color sensitivity done in the dark-room, because in the dark-room the pre-
exposure and surrounding field cannot be made of the brightness of the color. They
will always therefore exert an effect on the color impression. Moreover since the colors
all differ in brightness, this effect will be exerted in different amounts on the different
colors. That is, the amount of brightness added by after-image or contrast depends
upon the amount of brightness difference, respectively, between pre-exposure and color
and surrounding field and color. As stated above this amount, when working in the
dark-room, will be different for the different colors. For this reason and also because
even the same amount of brightness excitation acts with different degrees of strength
upon the excitation set up by the different colors, it is especially important that 110 work
on the comparative sensitivity of the retina to the different colors should be done in the
dark-room. It should be done in a light room of a constant intensity of illumination and
with pre-exposure and surrounding field in each case of the brightness of the color to be
used. In this way alone can all the factors which influence the sensitivit}' of the retina,
extraneous to the source of light, be eliminated.
FERREE: TESTS FOR THE EFFICIENCY OF THE EYE 43
colored and gray papers (e. g., the papers of the Hering series)
may be used on a color mixer.4 If, on the other hand, it is
desirable to use the light of the spectrum or the light transmitted
through standard niters, the colored light may be cut down to
the threshold value by means of a sectored disk, the sectors of
which should be covered with a gray of the brightness of the
color at or near its threshold value. (2) For brightness discrimi-
nation also the threshold or limen test is the most sensitive and
practical, but when made in a well-illuminated room, it becomes
in effect a test for a just noticeable difference. This test may be
performed at different points in the brightness scale, e. g., when
the standard is black, near mid-gray, or white. As before, disks
of standard papers may be used on the color-mixer, or the light
from a given source may be varied by means of a sectored disk.5
(3) Visual acuity tests of the Snellen type, especially when used
in work in which it is required to make successive tests on the same
person, are open to the following objections, (a) The judgment
is in terms of recognition. A letter may be recognized when it is
not seen clearly. In any judgment based on the recognition of
even a single letter, memory plays an important role. It is, so far
as the writer knows, impossible to standardize this memory fac-
tor and to obtain results strictly in terms of acuteness of vision.
(b) The test card is made up of quite a long series of letters.
As the test progresses the letters are memorized more and more
completely. It is practically impossible to eliminate this progres-
sive error when a number of successive judgments have to be
4 In making the tests with reflected light, two sets of disks are mounted on a color
mixer (a) an outer disk of gray of the brightness of the color to be used, and (b) an inner
disk made up of this gray and the disk of color. To the inner gray disk, varying pre-
portions of the color are added until the threshold value or just noticeable color is
obtained. To facilitate the judgment of just noticeable color, the inner disk of gray plus
color is compared with the outer disk of gray as a standard. Since both grays are of the
brightness of the color, the addition of the colored sector to the inner disk produces no
change of brightness either to confuse the judgment of noticeable color, or to affect the
intensity of the color excitation actually aroused. In getting the threshold value, the
method of ascending and descending series should be used, that is, beginning with equali-
ty, the variation is towards noticeable difference and beginning with a difference greater
than noticeable, the variation is towards equality. An average of the two sets of
results is taken for the threshold value.
5 When the test is made with reflected light two sets of disks, an outer and an inner,
are mounted on the color mixer. Each set is made up of one white and one black disk.
Both sets of disks are set at the point in the brightness scale from which the variation
towards white or black is to be made. One is kept constant and the other is changed
until the judgment different is given. In making the judgment the method of ascending
and decending series is used and the results are averaged for the difference limen. This
difference limen is taken as the measure of the observer's sensitivity to brightness or
white light.
44 TRANSACTIONS I. E. S. PART II
made as is the case before a final result is reached in any single
visual acuity test and as is especially the case when a number of
successive tests have to be given to the same person, which
happens in much of the work involved in the solution of the
problem here proposed. It might be supposed that the memoriza-
tion of the series could be broken up by using in each successive
judgment in a single test or in the successive tests, as the case
may be, cards having a different distribution of the letters in the
series. Considerable inconvenience would, however, be involved
in giving the tests in this way and besides no guarantee could be
had that each judgment would present the same degree of diffi-
culty. That is, the series is made up of similar and dissimilar
letters. The dissimilar letters can be distinguished from each
other with less difficulty than the similar. It is practically impos-
sible to distribute the letters so that the individual tests may be
equally rigorous. This objection can, of course, be eliminated
in part by a careful selection of the test letters, but not entirely
because a series of letters uniformly similar cannot be found,
(c) The Snellen series contains quite a large number of letters.
The eye is found to fatigue and vision to blur before the series
is completed. This introduces an error which it is practically
impossible to render constant.
All of the above objections were eliminated in the tests finally
adopted by us by changing the type of judgment and by making
the test object in one case two parallel vertical lines stamped I
mm. apart on a white card6 and in another the letters li printed
in small type.7 In using these cards the observer's acuity of
6 The card is mounted on a sliding carrier which runs on a track made of two meter
rods fastened end to end on a folding base. The base is mounted on adjustable stands
fastened to a table. When making the test the apparatus is so adjusted that the track
carrying the test card is just below and close to the observer's eye. In order that the
observer's head maybe held steady he is required to bite an impression of his teeth,
previously made and hardened in wax on a mouth-board, which is rigidly fastened by a
heavy iron rod and accessories to the.table supporting the track and carrier.
7 Besides the letters lithe writer would recommend the following figures as test objects.
I
(i) — ■ — . The test is to distinguish clearlv the dot at the center. A test object in the
I
shape of a cross has the advantage of affording a steady control of fixation. According
to photographic records of involuntan7 eye-movement, where a variety of fixation
objects has been used, the cross is found to give the best control of fixation.
I I
(2) ■ or ■ . In these figures also the test is to distinguish the dot clearly. The
II
former figure, however, is a little too complicated. There is both a tendency to lose the
dot and for the lines to run together on either side. A simpler criterion gives an easier
and safer judgment. Doubtless with a little effort other figures can be found possessing
still greater merit as test objects.
FERREE : TESTS FOR THE EFFICIENCY OF THE EYE 45
vision is determined by the distance at which he can just clearly
distinguish in every detail the two test objects. The results are
thus rendered directly in terms of clearness of vision, and there
are no progressive errors introduced by memory and fatigue.
We have good reason to believe that the brightness sensitivity,
color sensitivity, and visual acuity tests rendered sensitive and
adapted to our purpose in the manner described above will serve
as a measure of the general level of efficiency of the unfatigued
eye under different systems of illumination. For example, they
show considerable difference in result when the tests are given
under three types of lighting now in use: namely, systems of
direct lighting, systems of indirect lighting, and daylight. In each
of these cases, the intensity of the light falling on the test object
measured in foot-candles is kept the same. The tests can not,
however, be depended upon to show a loss of efficiency of the
eye as a result of three or four hours of work even under a very
unfavorable lighting system.
III. LOSS OF EFFICIENCY AS THE RESULT OF A
PERIOD OF WORK.
We have no reason to believe that the brightness and color
sensitivity tests have failed to show that the eye loses in efficiency
as the result of a period of work under an unfavorable lighting
system because of any fault in the tests. The tests used are the
product of several years of study by the writer of the sensitivity
of the eye to brightness and color and of the factors that in-
fluence this sensitivity. There is doubtless very little, if any, loss
of sensitivity during this length of time. In fact it is commonly
believed that the brightness and color processes are compensating
in nature. The case is quite different, however, with the conven-
tional visual acuity test, or even with the modification of it de-
scribed above. Although brightness and color sensitivity are fac-
tors influencing the visual acuity test, still in every case to which
it may be applied, it is predominantly a test of the refracting
mechanism of the eye and its muscular control. In fact our re-
sults for the tests of brightness and color sensitivity teach us that
when applied to the case in hand in which there has been no
change in the quality and intensity of the illumination or of the
refracting mechanism from the beginning to the close of work,
46 TRANSACTIONS I. E. S. — PART II
the results of the visual acuity test may be ascribed practically
entirely to changes in the muscular control of the refracting
mechanism, or at least to changes in the muscular control of the
eyes as a whole. s Now the visual acuity test, when it is confined
to a momentary judgment of clearness of vision, is not adapted
to show a loss in muscular efficiency because, although this
efficiency may have been lowered enormously, it may rise momen-
tarily under the spur of the test to its usual level, or at least to
the level obtaining at the beginning of work. Just as the runner
may, under the spur of his will, equal in the last lap of his course
the highest speed he has attained at any other point in the course ;
so may the flagging muscles of the eye be whipped up to their
normal power long enough to make the judgment required by the
visual acuity test. It was the feeling of all our observers that at
the close of work under the system of direct lighting installed in
our laboratory the eye had lost heavily in efficiency. A great deal
of discomfort was felt. The test was painful and was accom-
plished only with decided strain. Still the judgment could be
made apparently with as much accuracy as at the beginning of
work. But just as the runner finishing his course cannot long keep
up his extra burst of speed, so might we expect that the eye
cannot sustain its extra effort. This analogy led the writer
to continue the visual acuity test through an interval of time.
After considerable experimentation an interval of three minutes
was chosen as best suited for our purpose. Our surmise proved
to be correct. The fatigued eye cannot keep up its extra effort.
The results of the test showed an enormous loss of efficiency as
8 Before the writer would speak with full certaint3', however, that the retina
loses none of its power to function for color and brightness sensation during the above
stated period of work, he would feel it necessary to perform another kind of test for color
and brightness sensitivity. This test has been devised by him especially to meet the
needs of this problem. In this test the element of time is introduced. It is possible
that the retina may have lost in power to give color and brightness sensation as the result
of a period of work even when the conventional test based on a momentary judgment,
shows no loss of sensitivity. That is, it may be more susceptible to fatigue as the result
of the preceding work. To determine this, a fatigue test should be run at the beginning
and close of work. For color this may be done in two ways, (i) A given amount of
colored light may be used and the time required for the eye to become completely ex-
hausted or insensitive to this color may be determined. The difference in time required
for this amount of fatigue to take place at the beginning and at the close of work will
represent how much the retina has lost in its power to function forcolor. (2) Theexperi-
meni need not be continued until complete exhaustion takes place. The amount of
exhaustion that has taken place in a given interval of time can be measured. As before^
this can be done at the beginning and at the close of work and the results can be compared
to find out how much the retina has lost in power to give color sensation.
ferrEE: tests for the efficiency of the eye 47
the consequence of three hours of work under the system of
direct lighting, while in daylight practically no loss was shown.
In detail the test is as follows. When the observer is required
to look at the test card for three minutes, the test objects, even
when the eyes are fresh, are not seen clearly for the whole time.
The muscular effort required to keep the eyes adjusted for clear
vision cannot be sustained steadily for that length of time. The
test objects are seen alternately as clear and blurred. The time
they are seen clear and blurred is recorded on a rotating drum
upon which a line registering seconds is also run. From this
record the ratio of the time seen clear to the time seen blurred
is determined. This ratio may be fairly taken as a measure of
the efficiency of the eye at the time the test is taken. In applying
the test to our problem the record is taken at the beginning and
at the close of work, and the ratios of the time clear and the time
blurred are compared for the two cases to determine how much
the eye has lost in efficiency as a result of work. Two values
were chosen for the distance at which the test card was placed
from the eye: (a) the maximal distance at which the test objects
could be seen clearly in the momentary judgment, and (b) a
distance less than this. The latter distance was chosen because
for the maximal distance towards the close of the test, even when
the eyes were fresh, the value of the time blurred became, it was
thought, excessively high. Results for the two distances, there-
fore, give probably a fairer expression of the loss in efficiency
than for the one.
The problem dealing with loss of efficiency as the writer has
conceived it presents two phases. We may investigate (a)
whether the eye shows a loss of efficiency after three or four
hours of work under a given lighting system, and (b) whether
there is a progressive loss of efficiency in working several
months or years under a given lighting system. Only the first
part of this investigation has been attempted thus far in our
work and it has been undertaken, not so much for the purpose
of making an exhaustive study of loss of efficiency under
a given set of conditions, as it has been to get a sensitive and
practical method of detecting loss of efficiency. In order to
determine whether the method we have described is practical and
sufficiently sensitive for our purpose, tests should be made on a
5
48 TRANSACTIONS I. E. S. — PART II
large number of people under a wide range of lighting condi-
tions. We have not as yet made tests under a wide range of
lighting conditions. We have chosen rather to begin with three
broad types of illumination now in general use ; systems of direct
lighting, systems of indirect lighting, and daylight. Types based
upon the distribution of light have been selected because it has
seemed to the writer, both from his own work and from a survey
of the work done by others, that distribution or diffuseness of
light is the most important factor we have yet to deal with in
our search for conditions that give minimum loss of efficiency
and maximum comfort in seeing. The quality of the light and
its intensity at the source are already pretty well taken care of,
apparently at least better taken care of in general practise, rela-
tive to their importance to the eye, than is distribution. A
detailed report of our results will not be given in this paper.
The following results selected as typical from a large number
of observations are appended, however, to show how the effi-
ciency of the eye as measured by the above test falls off as the
result of three hours of work under a system of direct lighting
as compared with daylight.
The tests were conducted in a room 30.5 feet (9.29 m.) long,
22.3 feet (6.797 m0 wide and 9.5 feet (2.895 m-) high. The
daylight illumination came from six windows, all on one side
provided with thin white curtains to secure the necessary control.
The artificial lighting9 was accomplished by means of two rows
of fixtures of four fixtures each. Each row was 6 feet (1.828
m.) from the side wall and the fixtures were 6 feet apart. Each
fixture was supplied with two 16 candle-power carbon lamps
29 inches (0.736 m.) from the ceiling with a white porcelain
reflector 16 inches (0.406 m.) in diameter fastened directly above.
The daylight tests were made at 9 a. m. and 12 m. Between these
limiting times, the observer was required to read pages of type,
uniform in size, printed upon paper of uniform texture of sur-
face and of uniform reflecting power. The tests for the system
9 This room gave the impression of being brilliantly lighted. The writer was amazed
to find, however, that only 2.5 foot-candles of light were received on the test card placed
about midway between two of the rows of lights and midwa5' between two sets of fixtures.
The walls and ceiling of the room were of plaster, natural finish, and the floor of dark
tiling. Before our tests were taken, the walls and ceilings were painted white which
nearly doubled the light received on the test card.
FERREE : TESTS FOR THE EFFICIENCY OF THE EYE 49
of direct lighting were taken at 7 p. m. and 10 p. m. During the
interval intervening, the observer was required to read type of
the same size and printed on the same paper as was used in the
daylight work. The reading was done in each case at exactly
the same spot in the room as at which the tests were made. The
intensity of illumination was also in both cases made as nearly
equal as it was possible to do by methods now available.10 The
two tests were always given on successive days but one. In order
to guarantee that the observer's physical and optical condition
should be as nearly the same for the two tests as it was possible
to obtain, he was required to rest during the day immediately
preceding each test. Since the li test has proven to be the more
sensitive, results will be given for it alone in the following table.
Column 1 of this table gives the time of day at which the work
was done and the tests were made. Column 2 gives the type of
test. Column 3 gives the distance of the test card from the eye.
As stated earlier in the paper, two distances were used; one the
maximum at which the test object could be seen clearly, the other
a distance less than this. Division A of the table gives the results
for the former distance ; division B, for the latter. Columns 4
and 5 respectively, give the number of times the test object was
seen clear and unclear. Column 6 gives the number of seconds
in the three minutes that the test object was seen clear, and
10 In order to equalize the intensity of illumination, a method of measurement is
required. Two methods were used by us ; photometry, and a more delicate method
based upon the sensitivity of the peripheral retina to brightness contrast. In case of the
former, a Sharp-Millar portable photometer was used. The light falling upon the test
card was measured in foot-candles and was made equal for each type of lighting. Full
details of the latter method will not be given here. As stated above it is based upon the
extreme sensitivity of the peripheral retina to brightness contrast, especially to the
induction by a white screen. To apply the method, some given illumination is taken
as standard. The amount of black induced by a white campimeter screen upon a
15 mm. area of some medium gray, {e. g. Hering gray No. 14) at an excentricity of 25
deg. in the temporal meridian, is measured. This amount of contrast is taken as the
index of that illumination. To duplicate the illumination at any succeeding time, the
intensity is varied until the same amount of contrast is induced by the white screen on
the gray at the 25 deg. point, for the same observer. This method was devised in the
writer's laboratory and he has found by repeated trials that, although it is not so con-
venient for many of the purposes for which the photometric method is used, it is many
times more sensitive than the traditional photometric method. The Sharp-Millar
photometer, like other photometers, is insensitive for the determination of the illumina-
tion of a room by daylight. This is because the standard field illuminated by the
tungsten lamp is deep orange in color, while the comparison field illuminated by day-
light is clear white. This difference in color tone makes the judgment of brightness
equality difficult to make and renders the instrument extremely insensitive for daylight
work.
50 TRANSACTIONS I. E. S. — PART II
column 7 the number of seconds unclear. Column 8 gives the
ratio of the total time clear to the total time unclear. This ratio
as stated earlier in the paper expresses the efficiency of the eye
for clear seeing for an interval of three minutes at the time at
which the test was taken.
TABLE I.
Showing How the Eye Falls Off in Efficiency as the Result of Three Hours
of Work under a System of Direct Lighting as Compared with Daylight.
In Division A the Test Card is Put at the Maximal Distance at Which the
Test Object Could be Seen Clearly; in Division B, at a Distance Less than
This.11
Time
of
day
Test
Distance
of card
from eye
cm.
Number
of
times
clear
Number
of
times
unclear
Total
time
clear
sec.
Total
time
unclear
sec.
Total time clear
Total time unclear
A.
9 A. M.
li
102
15
15
105.6
78.4
1.4
12 M.
li
102
15
14
103. 1
76.9
i-33
7 P. M.
li
75
18
18
II9.7
60.3
1.98
IO P. M.
B.
9 A. M.
li
75
15
15
55-4
124.6
0.44
li
92
14
13
136.8
43-2
3.16
12 M.
li
92
12
12
134-9
45-i
2.99
7 P. M.
li
65
24
23
141.8
38.2
3-7
IO P. M.
li
65
17
17
75-5
104.5
0.72
11 It will be noticed in the table that the ratio total time seen clear -f- total time seen
unclear is smaller for the test both at the beginning and at the close of work in division
A where the maximal distance at which the test object could be seen was used, than in
division B where a distance less than this was used. This is just what should be expected
from the nature of the test. For it may be said that, within limits, the nearer the object
is to the eye the greater is the proportion of time it should be seen clearly ; and, con-
versely, the farther the object is from the eye the smaller is the proportion of time it
should be seen clearly. It will also be noticed that the ratio is slightly larger when the
tests are made under the system of direct lighting than when made under daylight. The
explanation of this, too, is found in terms of the distances that were chosen for the test
object. These distances, relative to the maximal distance, were chosen shorter for the
artificial light than for daylight. This was done because of the large falling off in the
ratio gotten for the test at the close of work under the artificial light. Had the first ofthe
two distances used in these tests, for example, been chosen as near to the maximal dis-
tance for the artificial light as it was for daylight, the result ofthe test made at the close
of work would have been that, after the first interval seen as clear, the test object would
have been seen unclearly during the remainder of the test. At first glance one might
be tempted to think that the difference in the scale of magnitude for the two ratios, is due
to some inequality in the intensity of the illumination that was given by the two systems
of lighting. It is obvious on reflection, however, that the intensity of illumination can
have little or nothing to do with the scale of magnitude of these ratios. The intensity of
the illumination influences the maximal distances at which the test object can be seen
clearly but the scale of magnitude of the ratio, time clear to time unclear must depend
primarily upon how near the distance chosen for the test object is to the maximal
distance. (This principle, it is obvious, does not affect the comparison of the ratios
obtained at the beginning and close of work under a given lighting system for the dis-
( Continued on following page. )
FERREE: TESTS FOR THE EFFICIENCY OF THE EYE
51
In order to give a typical representation in graphic form of
the effect of three hours of work on the efficiency of the eye in
daylight and under the system of direct lighting, estimated in
terms of the test we have described, the results of the above table
are given in the form of a curve. In constructing this curve the
length of time of work is plotted along the abscissa and ratio of
the time the test object is seen clear to the time unclear, is plotted
along the ordinate. Each one of the large squares along the
abscissa represents an hour of work, and along the ordinate an
integer of the ratio. Figure I shows the result of division A and
figure II for division B of the table. An inspection of these
\z.m
M.
mooRM.
i
7.00 RK
aooA.M.
Fig. I.— Curve for division A of the table. Showing how the eye falls off in efficiency
for three hours of work under a system of direct lighting as compared with daylight.
curves shows that the efficiency of the eye measured by the ratio
of the time the test objects are seen clear to the time seen unclear,
falls off rapidly for the system of direct lighting but scarcely at
all for daylight.
Although it has been the purpose of this paper merely to out-
tance, once it is chosen for that system, is kept the same or both tests). As further
proof that the difference in the intensity of illumination had nothing whatever to do
with this result, the intensity of illumination was carefully determined immediately
before and after these tests and, if the readings showed any inconstancy in the illumi-
nation, the results were discarded and new tests were made. The above explanation
should be borne in mind also in examining the curves plotted from the results of the table.
The curve for division B of the table, for example, begins at a higher point on the ordi-
nate than for division A ; and the curve for the artificial illumination starts at a higher
point than the curve for daylight.
It is scarcely necessary to point out that neither the scale of magnitude of ratio nor
the point at which the curve starts is of any considerable consequence for our work. The
important thing is not how large is the ratio at the beginning of work, but how much it
falls off as the result of work. In fact the magnitude of ratio need not be taken into ac-
count at all any further than that it chances to be a coincident result of a condition that
seems to render our test more sensitive. That is, our results seem to show that the ratio
falls off more when the distance chosen for the test object is not too near the maximal dis-
tance. In future work, therefore, more care should be taken probably than was exercised
in this preliminary study to choose the distances for the test object so that in case of each
lighting system employed they shall sustain the same ratio to their corresponding
maximal distances.
52
TRANSACTIONS I. E. S. — PART II
line and in part to demonstrate a set of tests, a word of dis-
cussion and interpretation of the results we have reported may
not be out of place here. Since the visual acuity test (given
under constant, quality, intensity and distribution of light) is a
test largely of the refractive mechanism of the eye and its mus-
cular control and since the refractive mechanism could not have
changed during three or four hours of work, the obvious indi-
cation of the above result is that the loss of efficiency sustained
by the eye in these experiments is a loss in muscular efficiency.
This conclusion is borne out also by the fact, stated earlier in
the paper, that the direct tests of the efficiency of the retina,
namely, the test for brightness and color sensitivity did not show
conclusively any loss.12 Moreover, the conlusion is in line with
current conception. In current theory the retina is considered
as a mechanism more or less compensating in its action, while
700 RM.
\
9.00 A.M.
\
s
I2.0C
)M.
IOloo
RM.
Fig. II.— Curve for division B of the table. Showing how the eye falls off in efficiency
as the result of three hours of work under a system of direct lighting as compared
with daylight.
the muscles of the eye are not so considered. The following
reasons are suggested why the muscles of the eye giving both
fixation and accommodation should be subjected to a greater
strain by the system of direct lighting than by daylight. ( i ) The
bright images of the electric bulbs falling on the peripheral retina
which is in a perpetual state of darkness adaptation as compared
with the central retina and is therefore extremely sensitive in its
reaction to such intensive stimuli, sets up a reflex tendency for
the eye to fixate them instead of the letters which the observer
is engaged in reading. (2) Likewise, a strong reflex tendency
to accommodate for these brilliant sources of light all at differ-
13 This statement is also subject to the foot-note appended to the earlier statement.
FERREE : TESTS FOR THE EFFICIENCY OF THE EYE 53
ent distances from each other and from the lettered page, is set
UP- (3) These brilliant images falling upon a part of the retina
that is not adapted to them causing as they do acute discomfort
in a very short period of time,13 doubtless induce spasmodic con-
tractions of the muscles which both disturb the clearness of
vision and greatly accentuate the fatiguing of the muscles. The
net result of all these causes is excessive muscular strain which
soon shows itself as a loss in power to do work. In the illumina-
tion of a room by daylight with a proper distribution of windows,
the situation is quite different. The field of vision contains no
bright sources of light to distract fixation and accommodation
and to cause spasmodic muscular disturbances, due to the action
of intensive light sources upon the dark adapted and sensitive
peripheral retina. In daylight the light waves have suffered
innumerable reflections and the light has become diffuse. The
field of vision is uniformly illuminated. The illumination of the
retina, therefore, falls off more or less uniformly from fovea to
periphery as it should in order to permit of fixation and accom-
modation for a given object with the minimum amount of strain.
It is not our purpose to contend in this report that distribution
is the only factor of importance in the illumination of a room.
The intensity and the quality of the illumination must also be
taken into consideration. To test the relative effect of these
factors upon the working power of the eye, records would have
to be taken when each was varied in turn and the other two
maintained constant. In the results shown in the above tables
the intensity alone was constant in the two cases. Both the
quality and the distribution were different in the direct lighting
system and the illumination by daylight. The difference in the
results obtained will have, therefore, to be attributed both to
13 There is no doubt in the writer's mind that the eye-discomfort experienced as the
result of work under an unfavorable system of lighting is not by any means all muscular.
The "sandiness" passing over into a stinging, stabbing pain which comes early, in the
experience of discomfort seems to be conjunctival. And while the retina itself is
apparently insensitive to pain from mechanical stimulation, still when exposed to a
source of light of a brilliancy to which it is not adapted," a painful reaction is produced
which can scarcely be considered muscular. For example, after confinement for some
time in a dark-room exposure to ordinary daylight is painful to the normal e3-e. That
this is not entirely muscular can be shown by the fact that a similar reaction is experi-
enced when the ciliary and iris muscles are paralyzed by atropine. The reaction is also
experienced by aphakial subjects whose lenses have been so long removed that muscular
atrophy must have taken place.
54 TRANSACTIONS I. E. S. — PART II
difference in the distribution and to difference in the quality of
the illumination. In our tests comparative of the systems of
direct and indirect lighting, the results of which will be reported
in a later paper, clear tungsten lamps will be used in both cases.
The intensity will be made the same and the quality of the light
will be approximately the same. The distribution or diffusion
alone will be different. Whatever difference in result we get in
these two cases can, therefore, with reasonable certainty be
attributed to the differences in the distribution of light.
With regard to the effect of varying the intensity of illumina-
tion, our results show nothing; with regard to the effect of varying
quality, nothing in isolation; and with regard to distribution, we
have data only for such differences as are found in the three
types of illumination now in general use. In later work, however,
the analysis along these lines will be completed. We hope on
the laboratory side, to make a systematic study of the effect of
wide ranges of variation of each of the factors in turn. It will
be comparatively easy, for example, to keep the intensity and
distribution constant and vary the quality, or to keep the quality
and distribution constant and vary the intensity. We hope in
addition, to supplement this work by testing the eyes of employ-
ees who work under a. given lighting system for several hours a
day, for evidences of a progressive loss of efficiency.
IV. A PRELIMINARY STUDY OF THE CAUSES OF
DISCOMFORT.
In addition to studying the conditions that give us maximal
efficiency, it is important to determine the lighting conditions
and eye factors that cause discomfort. In fact, it might well be
said that our problem in lighting at present is not so much how
to see better as it is how to see with more comfort and with less
damage to the general health on account of eye strain. Any
comparative study of the conditions producing discomfort neces-
sitates a means of estimating discomfort. It is obvious that the
core of the experience of discomfort is either a sensation or a
complex of sensations. As such, it should have a limen or
threshold value just as other sensations have; and just as we are
able in general to estimate sensitivity in terms of the threshold
value, so should we in this case be able to use the threshold
FERREE : TESTS FOR THE EFFICIENCY OF THE EYE 55
value in estimating the eye's sensitivity or liability to discomfort
under a given lighting condition. Threshold values are usually
determined by finding how much energy or intensity of a given
stimulus applied for a short interval of time is required to arouse
a just noticeable sensation. This form of procedure, however,
is not adapted to the needs of our problem. It is much better to
reverse the process and find how long the eye has to be exposed
to a stimulus of a given intensity to arouse just noticeable dis-
comfort. Our limen, then, becomes a time limen, and is meas-
ured in units of time instead of in units of intensity. In order
to determine whether the judgment of the limen of discomfort
can be made with certainty and to test in general the feasibility
of the method, the writer undertook to determine the compara-
tive sensitivity of the eye to discomfort when the source of light
was exposed in different parts of the field of vision. In order to
carry out this investigation, a 16 candle-power lamp was attached
to the arm of a perimeter in such a way that the end of the bulb
was always directed towards the observer's eye. The arm of the
perimeter could be shifted to any meridian in which it was desired
to work, and the lamp could be moved at will along the arm. It
was thus possible to expose the light at any point in the field of
vision that was desired. Working in this way, we have investi-
gated the effect of many types of variation of the distribution of
the light in the visual field, and it is our purpose to extend the
investigation as fast as possible to the variation of the other
factors. Of the variations we have made in the distribution of
the light in the field of vision, it will be necessary, however, in
order to illustrate the general method of working, to describe
only one, namely, the exposure of the source of light at different
points in the field of vision for one eye when fixation and accom-
modation were taken for a far point.
In carrying out the investigation, the following precautions
were observed, (a) It was found better to work in a room mod-
erately illuminated by a source of light behind the observer and
entirely concealed from him rather than in the dark. The inter-
vals of dark-adaptation between exposures in the dark-room
seemed to make the observer's eye too sensitive for our purpose.
This was especially true for certain parts of the peripheral retina.
6
56 TRANSACTIONS I. E. S. — PART II
In becoming supersensitive there was a tendency to become
erratically sensitive, (b) It was found that blinking serves as a
variable factor for the relief of discomfort and that the amount
of blinking must be made constant from test to test. This was
accomplished by having the observer blink at equal intervals dur-
ing the exposure, timing himself by the stroke of a metronome.
The interval most natural and suitable for this purpose was de-
termined for each observer separately, (c) All comparisons were
planned in series. For example, if it were desired to compare
the sensitivity of the temporal and nasal halves of the retina in a
given meridian, the exposure was first made at a given point in
one half and next at the corresponding point in the other half.
This was to guarantee that the eye should be as nearly in the
same condition with regard to progressive fatigue, etc., as was
possible. Further to safeguard against error in this regard ser-
ies were compared in which the exposures were repeated in the
reverse order, (d) An interval of recovery was allowed between
exposures. This interval had to be determined separately for
each observer and often had to be made different for the same
observer on different days. It was never changed, however,
during the course of an experiment, the results of which were to
be compared, (e) In order that the observer's head be held
rigidly in position during the exposure, he was required to bite
an impression of his teeth previously made and hardened in wax
on a mouthboard. When an exposure was to be made, the fixa-
tion was taken, the light turned on, and a signal was given by the
observer when a just noticeable discomfort was aroused, or, if
it was desired, when the different stages of discomfort were
reached. The judgment was found to present no especial diffi-
culty, and the method, when properly applied, to provide a feas-
ible means for comparing the sensitivity of the eye to discom-
fort under all the conditions to which we have been able thus far
to extend its application. In actual practise the method also
brings out an analysis of discomfort.
Discomfort seems to be a complex of three experiences, each of
which develops at a different time. When the light is turned on, we
have at once glare. This is a light sensation and though unpleas-
ant has no painful elements. Next comes a conjunctival sensation
FERREE : TESTS FOR THE EFFICIENCY OF THE EYE 57
which begins with what is commonly called "sandiness" and soon
passes over into a sharp, stinging, stabbing pain. Lastly there
comes what is probably a muscular discomfort, — a hurting and
aching in the ball of the eye which if the exposure is continued
long enough seems to radiate to the socket and the surrounding
regions of the face and head, the arch of the brow, the forehead,
the temples, etc. Details will not be given here of the compara-
tive sensitivity of different points of the retina to discomfort. It
will be sufficient to say, that the periphery of the retina is more
sensitive than the center; that the nasal half is in general more
sensitive than the temporal half and the upper half than the
lower half ; and that in passing from the center to the periphery
of the retina, the sensitivity is found first to increase then to de-
crease, becoming extremely little at the limits of the field of
vision. In the horizontal meridians both on the temporal and
nasal sides, maximal sensitivity is found around the 45 deg.
point. In the vertical meridians, maximal sensitivity seems to
be near the point 15 deg. below the horizontal. In a paper soon
to be published, a detailed statement and explanation of these
results will be given.
DISCUSSION.
Dr. H. E. Ives: This paper is well worth the while of pro-
fessional psychologists to study ; and it gives to the illuminating
engineer results which are extremely valuable.
When the illuminating engineer has the problem of producing
satisfactory results, he has two methods of doing so ; first, the
case system, in which he copies an illumination produced by
nature or invention, which has proved satisfactory by experience;
and he hopes to get the same result. But there are defects in
this method ; we are very apt to follow the example of the Chinese
who made motors by copying the imported ones even down to
the color of the paint on the casings and the scratches on the paint
We may do equally foolish things by slavish copying. That is in-
herent in the case system. Up to recently some of us have been of
the opinion that even with the defects of this case system we could
apply it to advantage, for instance by studying how nature pro-
duces her lighting schemes. But in order to make any great
58 TRANSACTIONS I. t. S. — PART II
progress we must deviate from what exists ; we must experiment
and invent. This necessitates some means of testing our results
and this process of experiment and test constitute the second pro-
cedure. Our Society has lately been interested in the physiological
side of illumination, but has been sadly handicapped by the lack of
significant tests — we have been dependent practically on labor-
iously acquired experience. One great object in adopting a
method of measurement is the saving of time. For instance,
suppose our only means of measuring voltage was by the duration
of physiological disturbances following an electric shock. In order
to duplicate a voltage which gives a shock whose after effects
last a day, we would require weeks or months of toil, because of
the time necessary to wait for the results of each experiment.
Suppose the first time we secured the desired voltage we had an
instrument known as a "voltmeter;" it would only take a minute
to determine that voltage. We want something for measuring
the effect of lighting systems which will enable us to get results
with a speed comparable with that of a voltmeter.
Various methods of test have been proposed and Dr. Ferree
has gone over all of these. He arrives at a conclusion which I
think it behooves us all to observe; namely, that these tests will
show what he calls "the general level or scale of visual efficiency,"
but they are practically useless as tests of the loss of visual
efficiency.
Here is a sentence which means a great deal "Just as a runner
may, under the spur of his will, equal in the last lap of his course
the highest speed he has attained at any other point in the course,
so may the flagging muscles of the eye be whipped up to their
normal power long enough to make the judgment required by the
visual acuity test."
Dr. Ferree here gives us the benefit of his point of view and
experience in these matters. In this paper he has recognized the
inefficiency of the methods now used. He realizes that we want
a test of the loss of visual efficiency. The eye may respond
momentarily, like the tired runner, and see the object as dis-
tinctly as before, but we know that it is not as efficient. Dr.
Ferree has devised a test in which is introduced a time element.
The observer views a visual acuity test object. When the
FERREE : TESTS FOR THE EFFICIENCY OF THE EYE 59
limit of visibility is found the observer is not allowed to rest,
because he will again after an interval get just as good results
as at first ; instead he presses a key as long as the detail is clear ;
then when the tired muscles flag and the object blurs, the finger
on the key is removed. At first it appears easy to see the detail
clearly, but pretty soon it is not so easy and one does not distin-
guish the chart so well. Very soon it becomes necessary to take the
finger off the key. Intervals of clear and blurred vision alter-
nate and at the end we have a ratio of the time the chart is dis-
tinguishable to the time when it is not.
Dr. Ferree has tried out daylight and a direct artificial
lighting system and we have here for the first time the results of
that test. They show what many of us have been sure of ;
that daylight does not decrease the efficiency nearly as much as
artificial lighting. On the fourteenth and fifteenth pages are two
charts showing by straight lines the falling off in efficiency which
occurs under artificial lighting as compared to daylight. Person-
ally, I think we should say "Eureka!"
I hope Dr. Ferree will proceed to standardize these tests and
tell us the best working distances and one thing and another.
As he is not here, I have tried to bring out the most important
points. He has given us a most valuable contribution, and I
hope before long we will be in a position to settle these questions
of light and dark walls by this method of test and not by
"Kilkennycat" discussion, which brings us nowhere.
I think we should do our best to aid Dr. Ferree to develop this
method of test to give us what now we can get only by experience.
I am aware that I have not done this paper justice, but I want to
express my appreciation of his work.
Mr. C. O. Bond: The American Medical Association is a
body whose conclusions as to the harmful or beneficial effects of
any types of illuminating installations will carry considerable
weight. They have discussed time after time how they were to
make the tests, and this paper has grown out of Dr. Ferree's
experiments, in the hope of placing in the hands of that Com-
mittee means of making the tests. We are extremely fortunate
in having the first public report of this method. The method is
under advisement by the Committee and I was present at one of
their meetings when Dr. Ferree brought a set of this apparatus
60 TRANSACTIONS I. E. S. — PART II
to Philadelphia and they made a test of it. Two or three of the
doctors present were very much impressed with it. I think, even
if it does not succeed as it now stands, perhaps here is the germ
of the best possible method of test.
Dr. C. E. FerrEE (communicated in reply) : I can express
only great appreciation of the interest that the men who have pre-
ceded me have taken in our work. The problem is extremely in-
teresting to me and I hope we have here a vulnerable point of
attack. Once we have procured a successful method of measur-
ing the effect of different lighting systems on the eye, a broad field
of application opens out before us. We not only can find out
what are the favorable and what are the unfavorable features in a
lighting system, but we can no doubt, as may be inferred from Dr.
Ives' discussion, test out and perfect a lighting system, so far as
its effect on the eye is concerned, before we put it on the market.
This latter point is a good one, I think, and I thank Dr. Ives for
the suggestion.* I feel that Dr. Ives' perspective and practical
grasp of the situation is a distinct contribution to the paper.
We are very much handicapped at present for funds by means
of which to carry on this work. In the first place apparatus and
models of lighting systems are required for the work on the
laboratory side. Trained assistants are also needed to help out
with the details of the work. Further, to verify and enlarge
the work done in miniature in the laboratory, we should test the
eyes of employees working under established lighting systems
and in the surroundings in which these systems have to operate.
All of this takes time and money, also entrance into commercial
concerns. In all of these regards we need the help and influence
of the Illuminating Engineering Society.
This work, I suppose, could be done spontaneously and sporadi-
cally here and there as the insight and inclination of various men
may direct. But in the beginning, at least, I do not think
it should be scattered. Until launched and safely moving, it
should be done under common supervison.
* The general idea that over and above its application to abstract investigation the
test may have an application in the daily work of the lighting engineer has come to the
writer by suggestion from the engineers themselves. Mr. Cravath, for example, has re-
cently pointed out that the test should be of advantage in making the actual installation
of a lighting system. The writer would suggest in addition that it may further be of ser-
vice in determining the effect of different kinds of type and paper on the efficiency of the
eye: also the effect of different kinds of desk lighting, etc. In short, it is obvious that the
usefulness of such a test is limited along these lines only by its sensitivity.
TRANSACTIONS
OF THE
Illuminating Engineering Society
Published on the 2Sth of each month, except during July, August, and September, by the
ILLUMINATING ENGINEERING SOCIETY
General Offices.- 29 West Thirty-Ninth Street. New York
Vol. VIM
FEBRUARY, 1913
No. 2
Index for Volume VII.
The index for Volume VII (1912) of
the Transactions is mailed with this
number.
Council Notes.
The council held a regular meeting in
the general offices of the society, 29
West 39th Street, New York, February
14. 1913. Those in attendance were:
Preston S. Millar, president; George S.
Barrows, Louis Bell, C. O. Bond, J. R.
Cravath, Joseph D. Israel, general sec-
retary; V. R. Lansingh, Norman Mac-
beth, L,. B. Marks, treasurer; W. Cullen
Morris, C. J. Russell and W. J. Serrill.
A monthly report on the membership
and the receipts and expenses was re-
ceived from the general secretary. The
number of members, counting the
applications and resignations presented
at the meeting, was said to total 1,325.
Eleven applicants were elected mem-
bers. Their names appear on another
page of this number.
Reports on section activities were re-
ceived from Vice-presidents J. R.
Cravath (Chicago), Howard S. Evans
(Pittsburgh). J-. W. Cowles (Boston),
\V. J. Serrill (Philadelphia).
A tentative report on proposed work
was received from the chairman of the
section development committee. It was
stated that the first meeting of the com-
mittee was to be held February 15 and
that a more definite report would be
presented later.
An oral report was received from the
papers committee. The arrangements
for the papers program of the next con-
vention were discussed briefly. It was
said that it might be well to arrange
the next convention program somewhat
as follows : the first day to be devoted
to society affairs; the second day to
the presentation of technical papers;
the third day to the reading of papers
of a commercial character; and the last
day to a series of lectures and talks on
the architectural and decorative aspects
of illuminating engineering. It was
understood that a definite report would
be received from the papers committee
at the March meeting of the council.
In a report received from the com-
mittee on editing and publication it was
stated that beginning with the 1913
issue of the Transactions an inexpen-
sive mat surface paper for the ordinary
text matter and line illustrations and
a coated paper for the photographic
reproductions would be used ; each issue
of the Transactions will be printed in
two parts; Part I will be devoted to
council and section notes and news
items; Part II will include papers, dis-
cussions and important committee re-
ports. The committee also asked for
an appropriation of $20.00 to print two
or three hundred copies of a guide
TRANSACTIONS I. K. S. — PART I
setting forth the character and style of
papers for presentation at meetings of
the society and for publication in the
Transactions. The appropriation was
authorized.
Progress reports were received from
the sustaining membership committee,
the committee on new membership and
the committee on reciprocal relations
with other societies.
A tentative report from the commit-
tee on glare from reflecting surfaces
outlined the work of the committee for
the present year. In its work the com-
mittee will give special attention to
school officials and school book com-
panies. The co-operation of other
societies will be solicited. Information
regarding paper making and the pub-
lishing art will be continually sought.
Encouragement will be given to manu-
facturers to produce a dull finished
paper which will reproduce half-tones
satisfactorily. The committee also be-
lieves that it is possible to collect suffi-
cient data regarding the use of black-
boards, polished desk tops, glazed paper,
etc., to form a paper of sufficient value
to find a place on the program of the
next convention of the society. Briefly
speaking, the committee will make every
effort toward the elimination of polished
surfaces, but for the present it will
confine its efforts to the public schools.
The finance committee reported that
it had approved for payment vouchers
Nos. 1170 and 1173, inclusive, and 1175
to 1209, inclusive, aggregating $759.02.
The factory lighting legislation com-
mittee reported that the recommenda-
tions which it had submitted to the
New York Factory Investigating Com-
mission pertaining to bill No. 18 having
to do with the lighting of factories and
work rooms had been incorporated in
the revised bill of the commission.
An oral report was received from
the committee on illumination primer.
The committee asked that the council
authorize the publication of a large edi-
tion of the primer to fill any orders that
may be received, and that permission
be given to make a few changes in the
primer. The council authorized the
committee to make such changes as it
may deem necessary, and the printing
of such editions of the primer as may
be required from time to time. It was
understood that these editions would
be gotten out by the general office.
A report was received from Mr. C. A.
Littlefield, chairman of the committee
of arrangements for the 1913 annual
meeting. It was resolved to extend to
Mr. Littlefield and the members of his
committee, on behalf of the society, a
very hearty vote of thanks for the
excellent and successful meeting and
dinner which they had arranged.
Upon receipt of a report from Mr.
V. R. Lansingh, chairman of a pre-
liminary committee, it was resolved that
the president appoint a committee to
foster the establishment, maintenance
and development of courses in illumi-
nating engineering in colleges and uni-
versities. The president appointed the
following committee, which is to be
known as the committee on collegiate
education : V. R. Lansingh, chairman ;
Prof. Henry B. Dates, secretary; Df.
H. E. Ives.
The president was requested to ap-
point a committee to consider and re-
port to the council upon the advisability
(1) of holding the 1915 annual con-
vention in San Francisco, and (2) of
endeavoring to arrange for that time
and place a joint meeting of the several
illuminating engineering societies or a
meeting of the proposed International
Commission on Illumination, or both.
TRANSACTIONS I. E. S. — PART I
The president was also requested to
appoint a committee on popular lectures,
said committee to be asked to promul-
gate a plan for preparation of popular
lectures on various classes of lighting
installations (as factory, store, resi-
dence, etc.) and, after plan has been
approved by the council, to undertake
the preparation of such lectures, either
directly or through sub-committees.
It was resolved that the council of
the Illuminating Engineering Society
takes pleasure in accepting the invita-
tion of the American Gas Institute to
join in a session on illumination at
Richmond Ya., on October 16, 1913.
New Members.
The following applicants were elected
members of the society at a meeting of
the council held February 14, 1913 :
Burrows, S. B.
Public Service Electric Co., Newark,
N. J.
Clinch, Edward S., Jr.
Lord Electric Co., 105 West 40th
Street, New York, N. Y.
Frey, Arthur C.
Cadet Engineer, United Gas Im-
provement Co.. Broad and Arch
Streets, Philadelphia, Pa.
Gudge, B. J.
Electrical Engineer. Westinghouse
Elec. & Mfg. Co.. East Pittsburgh,
Pa.
Higgixs, Warren Snedek.
Instructor in Electrical Engineering,
Lafayette College, Easton, Pa.
Horner, Harry Archer.
Electrical Engineer. New York
Shipbuilding Co., Camden, N. J.
Kelsey, Fenton P.
Vice-President and Editor. Gas
Record, Chicago, 111.
Malia, James P.
Chief Electrician, Armour & Co.,
Union Stock Yards. Chicago, 111.
McCulloch, Fred. H.
Treasurer, Electric Supply & Fix-
ture Co., 123 East Washington
Street, Fort Wayne. Ind.
Skinner. Ross H.
Assistant to General Contracting
Agent, Consolidated Gas Co., 435
Sixth Avenue, Pittsburgh, Pa.
Stafford, Raymoxd W.
Assistant Foreman, The New York
Edison Company, 117 West 39th
Street, New York City.
Section Activities.
CHICAGO SECTIOX
A joint meeting of engineering, archi-
tectural and ophthalmological societies
was held at the Republican House in
Milwaukee, February 22, 1913. The
following societies participated :
Chicago Section of the Illuminating
Engineering Society.
Engineering Society of Wisconsin.
Milwaukee Company Section of the
National Electric Light Association.
Milwaukee Electrical League.
Milwaukee Engineering Society.
Milwaukee Oto-ophthalmic Club.
Milwaukee Section of the American
Chemical Society.
Milwaukee Section of the American
Institute of Electrical Engineers.
Madison Section of the American
Institute of Electrical Engineers.
Wisconsin Chapter of the American
Institute of Architects.
The program presented was as
follows;
"Light and Art," by Mr. M. Luckiesh,
engineer, National Electric Lamp Asso-
ciation, Cleveland, O.
Discussion of "Ocular Comfort and
its Relation to Glare from Reflecting
TRANSACTIONS I. E. S. — PART I
Surfaces," by Mr. F. A. Vaughn, con-
sulting engineer, and Dr. Nelson M.
Black, ophthalmologist, Milwaukee,
Wis.
"A Photometer Screen for Use in
Tests of Street Illumination," by Prof.
Arthur H. Ford, State University of
Iowa, Iowa City.
"The Influence of Colored Surround-
ings on the Color of the Useful Light,"
by Mr. M. Luckiesh, engineer, National
Electric Lamp Association, Cleveland,
O.
These papers and discussions will
appear in later issues of the Trans-
actions. The meeting was well at-
tended, some 30 or 40 Chicago men,
besides members from Madison, Mil-
waukee and other Wisconsin and Michi-
gan points being present.
NEW ENGLAND SECTION
The New England section held a joint
meeting with the Boston section of the
American Institute of Electrical Engi-
neers in the auditorium of the Boston
Edison Building, 39 Boylston Street,
Boston, February 17, 1913. The follow-
ing papers were read :
"Street Lighting with Ornamental
Luminous Arc Lamps" by C. A. B. Hal-
vorson, Jr., of the General Electric Com-
pany, West Lynn, Mass.
"The Theory of Mercury- Vapor Ap-
paratus" by P. H. Thomas, consulting
engineer, New York.
"Flame Carbon Arc Lamps" by W. A.
Darrah of the Westinghouse Electric
& Manufacturing Company, East Pitts-
burgh, Pa.
The first two of these papers appear
in this issue of the Transactions. The
last one will be published later.
NEW YORK SECTION
A joint meeting of the New York
section of the I. E. S. and the Munici-
pal Art Society was held in the National
Arts Club, 119 East 19th Street, New
York, February 12, 1913. The topic of
the evening was "Municipal Lighting."
Addresses were made by C. F. Lacombe,
chief engineer of the Department of
Water Supply, Gas and Electricity of
New York City; Charles Roland Lamb,
ecclesiastical architect; William Wentz,
vice-president, O. J. Gude Company, and
Arthur Williams, general inspector, The
New York Edison Company. A dinner
preceded the meeting and was attended
by guests and members of both the
club and the society.
The program of meetings for the re-
mainder of the season is as follows :
March 13 — Joint meeting with the
American Society of Mechanical Engi-
neers in the United Engineering So-
cieties Building, 29 West 39th Street,
New York. Mr. Ward Harrison of the
National Electric Lamp Association will
present a paper on "Industrial Lighting."
April 9 — This meeting will probably
be held in the United Engineering
Societies Building. Mr. M. Luckiesh of
the National Electric Lamp Association
will present a paper on "Light and Art."
A paper on "Phosphorescence and
Fluorescence" is also scheduled. This
meeting should be an unusually inter-
esting one.
May 8 — A talk on theater lighting by
Mr. Bassett Jones, Jr., at the Clymer
Street Theater, Brooklyn. During the
past year Mr. Jones has conducted a
great deal of experimental work in
theater illumination particularly in the
production of stage effects. The mem-
bers of the New York chapter of the
American Institute of Architects will be
invited to attend this meeting. Admis-
sion will be by card.
June 8 — It is planned to have a joint
meeting and outing of all the engineer-
ing societies in New York.
TRANSACTIONS I. E. S. — PART I
PHILADELPHIA SECTION
The February meeting of the Phila-
delphia section of the Illuminating En-
gineering Society was held on Feb-
ruary 21, 1913, at the Engineers' Club,
1317 Spruce Street, Philadelphia. At
the dinner preceding the meeting, which
was held in the dining-room of the
Engineers' Club, 46 members and guests
were present. One hundred and sev-
enty-five members including a number
of architects and ophthalmologists were
present.
Mr. P. S. Millar, president of the
society, presented a paper on "Some
Phases of .the Illumination of Inte-
riors." The subject was discussed by
Messrs. Perot, Bond, Regar, Hare,
Dickey, Swanfeld. Israel, Prof. Hoadley
and Dr. Crampton.
PITTSBURGH SECTION
A meeting of the Pittsburgh section
was held in the hall of the Engineers'
Society of Western Pennsylvania, Oliver
Building Pittsburgh, February 21, 1913.
A paper on gas illumination by Mr.
S. B. Stewart of the Consolidated Gas
Company was presented.
The program of meetings for the rest
of the season is as follows :
March — "Moving Picture Lanterns
from the Central Station Point of
View" by J. F. Martin.
April — "Railroad Car Lighting" by
J. L. Minick.
May — "Physiological Aspects of Il-
lumination" by W. E. Reed.
June — Announcement will be made
later.
to
TRANSACTIONS
OF THE
Illuminating
Engineering Society
FEBRUARY, 1913
PART II
Papers, Discussions and Reports
[ FEBRUARY, 1913 ]
CONTENTS - PART II
The Influence of Colored Surroundings on the Color of the
Useful Light. By M. Luckiesh 61
Theory of Mercury-vapor Apparatus. By Percy H. Thomas 75
Street Lighting with Ornamental Luminous Arc Lamps. By
C. A. B. Halvorson, Jr SS
THE INFLUENCE OF COLORED SURROUNDINGS ON
THE COLOR OF THE USEFUL LIGHT.*
BY M. LUCKIESH.
Synopsis : In the course of the investigation reported in this paper,
theoretical computations and also actual colorimeter measurements were made
to determine the magnitude of the influence of colored surroundings on the
color of the useful light. Much attention has heretofore been given to the colo r
value of illuminants, when after all the color value of the useful light is per-
haps of greater interest to the user. It was suspected that especially in in-
direct lighting the colored walls and ceiling, even though nearly white,
would effect appreciable color changes in the incident light. Calculations
given in this paper show that only a moderately yellow paper is required to
change the color of tungsten light to that of a carbon filament lamp. Both
direct and indirect lighting were considered in the computations ; the results
obtained are plotted in several ways.
A miniature room was fitted with a tungsten unit in the middle of the
ceiling. The walls and ceiling were covered with combinations of green-
yellow, and white paper. Colorimeter readings were made with direct and
indirect lighting. These are shown in a table and in a color-triangle. Re-
flection coefficients are briefly discussed and measurements presented for
several papers illuminated by various illuminants. These measurements
were made with a flicker photometer.
The color value of illuminants has been a subject of consider-
able interest during the last few years. The light from the
modern metallic filament lamps has been lauded by many because
of its nearer approach to daylight than many of the other common
illuminants, while a few have expressed favor for the "softer"
yellow light of the carbon filament lamp. Relatively the color
values of common illuminants differ considerably. When these
illuminants are used amid colored surroundings one is led to
suspect that considerable change in the color value of the useful
light might take place especially with indirect and semi-indirect
lighting. Of course great color changes can be effected by using
colored reflecting surfaces, the amount of alteration in color
depending upon conditions. For instance, there is no theoretical
reason why the light from ordinary illuminants cannot be altered
by reflection from colored surfaces to accurately match "average
1 A paper read at a meeting of the Chicago section of the Illuminating: Engineering
Society, in Milwaukee, Wis., February 22, 1913.
62 TRANSACTIONS I. E. S. — PART II
daylight." In nature considerable change takes place in the color
of sunlight on being reflected from the earth and especially from
green foliage. This has been shown by noting in this reflected
light the absorption bands of chlorophyl which is the green color-
ing matter of foliage. Measurements at sea or over a sufficiently
large snOw field in winter do not show these peculiarities. Mr.
G. S. Merrill1 has measured the color values of daylight on the
working plane in a room after part of it had been reflected from
the colored surroundings. The interior measurements were made
on clear and cloudy days. They showed considerable alteration
in the color of outdoor average daylight.
In direct artificial lighting much less light reaches the working
plane via the walls and ceiling than in semi-indirect or indirect
lighting. Obviously in the latter system all the light which is
diffusely reflected by colored walls or ceilings is altered in color.
When the reflection coefficient is small the influence of the colored
paper is small, especially in direct lighting owing to the small
amount of the colored light which is added to the direct light.
In such cases only about two reflections need be considered, as
there is little light remaining unabsorbed after it has suffered two
reflections. When the reflection coefficient of the surroundings
is large many more reflections must be taken into account as will
be shown later.
Assume an illuminant radiating equal amounts of monochro-
matic red, green and blue light which might be represented
relatively as
RGB
IOO IOO IOO
Further assume that this light source is placed in the center of
an Ulbricht sphere, the walls of which are covered with perfectly
diffusing green paper. The reflection coefficients of this paper
for the particular illuminant are assumed to be
R g b ,
25.2 47-2 27.6
The light received by the green paper is reflected an infinite
number of times. If the walls of the sphere are temporarily sup-
1 Proc. American Institute of Electrical Engineers, p. 1726, 1910.
LUCKIESH : INFLUENCE OF COLORED SURROUNDINGS 63
posed to be white and if N is the reflection coefficient of the
paper then the total light falling on the walls will be
Q = Q' + NQ' + N'Q' + N8Q' + =-- y^- (1)
where Q = total light falling on the walls,
Q' = direct light from the light source falling on the
walls.
Color values of a paper are determined by measuring the color
value of the light after it has suffered one reflection. A reflec-
tion coefficient of 33//i per cent, was assumed for the green paper
for this particular illuminant. The coefficient of reflection of
the paper can vary between wide limits without any change in
the color values.
Based on the foregoing assumptions, the reflection coefficient
of this paper for the monochromatic red light is 25.2 per cent,
of the original 100 units; 47.2 per cent, of the total 100 units of
green light; 27.6 per cent, for the total 100 units of blue light.
For this case the total red, green and blue components in the
light incident on the wall paper after an infinite number of reflec-
tions will be
QR = Q'R+NRQ'R+NK?Q'R+N|iO'R+ . . . = Q\T (2)
Qo = Q'c + NcQ'c + Ng-Q'g + Ng3Q'g + • • • = t ^V (3)
I
— NR
Q'c
I
-NG
Q'b
Qb = Q'b + NBQ'B + NJQ'b + N|Q'B + . . . = t ^^ (4)
and
Q = Qr + Qo + Qb = total light on walls (5)
Q' = Q'r + Q'c + Q'b = total direct light on walls (6)
NR, NG, NB are respectively the reflection coefficients for the
monochromatic red, green and blue light's.
NRQ'R, NGQ'G, NbO'b are the color values of the wall paper as
determined by the colorimeter under the light Q\
It is interesting to make some calculations in this particular
case.
64
TRANSACTIONS I. E. S. — PART II
TABLE I.— Computations According to Equations (2), (3) and (4),
Showing the Changes Produced in the Light from a Special
"White Source" by Successive Reflections from a Certain
Green Paper a.
The term in
Values
Values reduced for r.
color trianj
R G
lotting in
(2), (3) and (4) R
G
B
B
0/
100
IOO
IOO
33-3
33-3
33-3 a
NQ/.--
25.20
47.20
27.60
25.2
47-2
27.6 b
N2Q/ • ■ •
6-35
22.30
7.62
17-5
61.5
21.0 c
N3Q/.-.
1.60
IO.45
2.IO
n-3
73-9
14.8 d
NV-
0.40
4-93
O.58
6.8
83.4
4.8 e
N50/ • ■ ■
0. 10
2-33
O.I6
3-8
90.0
6.2 f
N6Q/ • • •
0,03
1. 10
O.04
2.6
94.0
3-4 g
N7Q/---
0.52
O.OI
N8Q/ • ■ •
0.25
N9Q/ • • •
0.12
The reduced values a, b, c, etc., are plotted in fig. 1. Theoret-
ically these values should be plotted in the color pyramid as they
Fig. 1.— Theoretical computations of the changes taking place in the color of light
under various conditions and after undergoing various reflections
are plotted in trilinear coordinates.
are of unequal luminosity but for clearness they are plotted in
one plane — the color triangle — which method shows their rela-
luckiesh: influence of colored surroundings
tive positions. It will be noted how the color of the light
approaches saturated green as the number of reflections increases.
In fig. 2 are plotted the actual luminosity values of the red, green
and blue components in the light after it has undergone various
reflections. The full lines represent the magnitude of the com-
ponents when the reflection coefficient of the paper is ZVA Per
cent, for the illuminant used. The dotted lines show the rapid
decrease in the values with a paper of 10 per cent, reflection
coefficient as compared with one of 335/3 per cent, coefficient. This
data of course is not truly represented by a curve because there
100
90
80
</>
560
t—
z
£50
J40
UJ
ac
30
20
10
i\
in 1 \
1 vU \
' \\\ \
1 \\\ \
1 \\\ \
|
1 v >
1
v \ '*
\ \
\
^R^N
- -~r
s5s**>i~-
___
D
•
> ;
'
1
> (
NUMBER 0FREFtECTIONS,6REEN PAPER
Fig. 2.— The percentage of the original light remaining after it has
suffered various reflections.
cannot be fractional parts of a reflection. The various points
corresponding to different reflections are connected by the curves
merely as a graphical means of representing the data. In fig. 3
are plotted the relative amounts of the red, green and blue com-
ponents in the light after it has suffered various reflections. The
sum of the ordinates at any particular reflection equals 100 per
cent. Here the rapid approach toward a pure green is shown
after the light has undergone several reflections.
Turning again to fig. 1 and Table I some more interesting cal-
culations are possible. First let us discard the direct light from
the calculations. Adding the light of the first and second reflec-
tions in columns 2. 3 and 4 and making the sum of the three
66
TRANSACTIONS I. E. S. — PART II
components equal to ioo we have the quality of the light reflected
once from the paper but diluted with the light of the second
reflection. This latter amount is small but more greenish in color.
This value is plotted with a cross and numbered 2, which indi-
cates the sum of two reflections. The sum of three, four and
five reflections are also plotted as crosses and labeled consistently.
The final color of the light after undergoing an infinite number
of reflections (and absorptions) is found by summating the
series in equations 2, 3 and 4 and subtracting in each case the
direct component. This value for indirect lighting might be
100
so
£0
70
60
50
30
£0
10
°<
e""--4/
B"\
R--'
)
r
NUM
3ER0FREE
5 i
.ECTIONS.G
REEN PAPE
> 6
R
Fig. 3. — The percentage of the three components in the light after undergoing
various reflections frorn green paper.
found by measuring the color of a white surface screened from
the light source. It is plotted as °o .
Next the effect of adding a constant amount of direct light Q'
to the reflected light is shown by the circles. Circle number i'
shows the color of the direct light as diluted by the first reflec-
tion. Circles 2' ', 3', 4', 6" indicate the change in the color of the
total light as more reflections are considered. The color of the
total light after it has undergone an infinite number of reflections
is found by summating each series in equations 2, 3 and 4, and
making the sum of the three values equal to 100. It is plotted
as 00 '.
LUCKIESH : INFLUENCE OF COLORED SURROUNDINGS 67
After purely theoretical computations made only for the pur-
pose of showing the order of magnitude of the possible color
changes, it becomes of interest to make some observations
with an Ives colorimeter and in a room papered with various
colored wall papers. With actual colored papers as used on
walls there is more or less specular reflection. The light which
is specularly reflected does not necessarily undergo the same
change in its color as in the case of diffuse reflection. This fact
must be considered when using the colorimeter for determining
the color of papers and other things. If the paper is placed at
such an angle as to regularly reflect the light from its illuminant
into the colorimeter obviously the true color of the paper will
not be obtained. Mere observation indicates that more likely the
colorimeter readings will more nearly represent the color of the
illuminant than that of the paper.
It is impractical to use an actual room in these experiments
owing to the great amount of light required for the colorimeter
readings and also the large surfaces which must be re-papered.
Besides the results would only hold for that particular room, wall
paper, etc. This of course is true with any arrangement of
apparatus. The object of these measurements was merely to
show the possible magnitude of the color changes in the illumi-
nant after it had suffered various reflections and had reached the
working plane. A cubical box, four feet on a side, was arranged
with a single fixture in the center of the ceiling. The light source
was a 500-watt tungsten lamp. This source was used directly
and indirectly. With the direct system no reflector was used, of
course permitting considerably more than the usual percentage of
light flux to reach the walls and ceiling. Green, yellow and white
papers were used. The green and yellow papers were selected
from regular wall paper stock and were quite unsaturated colors.
The white paper used was blotting paper. All colorimeter read-
ings were made with the photometric field of constant brightness.
The colorimeter readings under various conditions are shown
in Table II. It will be noted that the readings for the tungsten
lamp are shown as being equal for the red, green and blue. This
course was considered legitimate for several reasons. First it was
found impossible to make the readings comparable with data
68 TRANSACTIONS I. £. S. — PART II
heretofore published by others without entailing a vast amount of
work which was considered unwarranted owing to the fact that
the cases are not general. Further the chief object was to show
only the order of magnitude of the color-changes compared with
the magnitude of the color difference between the carbon and
tungsten incandescent lamps. The results are plotted in trilinear
co-ordinates in fig. 4. It is interesting to note that 4, 5, 6, 7 show
a gradual change from yellow toward "white" as would be ex-
pected in going from indirectly lighted yellow walls and ceiling to
a white ceiling and direct lighting. These results are consistent
. with the theoretical computations in the first part of the paper.
The color values of the carbon lamps (shown as 2 and 3 in
Table II) are relative to the tungsten lamp and are given in
order to illustrate the magnitude of the change in the color of
TABLE II. — Colorimeter Measurements in a Miniature
Room under Various Conditions.
RGB
1 Tungsten lamp 33.3 33.3 33.3
2 Carbon lamp, 3.1 watts per candle 38.7 34.7 26.6
3 Carbon lamp, 4 watts per candle 43.0 33.7 23.3
4 Yellow walls and ceiling, indirect 53.1 37.0 9.9
5 Yellow walls and ceiling, direct 47.6 35.7 16.7
6 Yellow walls and white ceiling, indirect. • • 43.2 35.5 21.3
7 Yellow walls and white ceiling, direct 42.1 35.3 27.6
8 Yellow paper . • • • 43.6 3S.4 18.0
9 Green paper 34.5 39.8 25.7
10 Green ceiling and green walls, indirect- •• • 35.9 43.6 17.5
11 Green ceiling and green walls, direct 36.3 37.8 25.9
12 Green walls and yellow ceiling, indirect . ■• 48.2 42.6 9.2
13 Green walls and yellow ceiling, direct 39.8 39.6 20.6
14 Green walls and white ceiling, indirect 36.6 34.1 29.3
15 Green walls and white ceiling, direct 35.2 34.6 30.2
the tungsten light due to colored surroundings. The color values
of the yellow and green papers used in the box are given at 8
and 9 and in the table. These were determined with the colo-
rimeter and show the color of the tungsten light after being
once reflected from the paper.
An instrument for observing the change in the color of incan-
descent lamp light due to colored surroundings is easily made.
A box 16 inches long and 4 inches square contains a lamp of
the same type and voltage as used in the room in question. This
LUCKIESH : INFLUENCE OF COLORED SURROUNDINGS 69
lamp is fastened in one end of the box and illuminates a piece
of opal glass. This portion of the box should have a white coat-
ing on its interior walls. Variation in the brightness of the glass
is obtained by means of a diaphragm. In an adjacent compart-
ment is a mirror inclined at 45 deg. to the vertical. Above this
is a ground opal glass which receives the light just as it is
received on the working plane. By means of the mirror this glass
is seen adjacent to the glass illuminated by the electric lamp.
When a brightness match is made the change in the color of the
illuminant is observable. Such a box was made and observable
changes in the color of the illuminant due to colored surroundings
Fig. 4.— Results obtained in a small room (plotted from Table II.).
were noted even in extreme cases of direct lighting. A modifica-
tion can be made by using in place of the electric lamp a ver-
tical tube which admits light to the first opal glass from the
illuminant only in much the same manner as the Sharp-Miller
photometer is used for candle-power measurements. This latter
arrangement could be used with all kinds of illuminants, but
would be useless in indirect lighting where perhaps the greatest
color changes are found.
Although it is well known that the coefficient of reflection of a
surface is not the same for all illuminants unless that surface is
white, it may not be out of place to discuss reflection coefficients
70 TRANSACTIONS I. E. S. — PART II
briefly. With the ever increasing diversity of types of light
sources and refinement of illuminating engineering procedure it
may soon become desirable when giving reflection coefficients to
also include in the statement the light source for which the coeffi-
cient was determined. To illustrate this point some relative
reflection coefficients of various colored papers of high saturation
are given in Table III. These were determined only relatively by
measuring the relative brightnesses in a fixed direction while the
specimen was illuminated successively by various illuminants at
hand. The brightness measurements were made by means of a
flicker photometer, the comparison standard being a white matt
surface (magnesia) illuminated by means of a tungsten lamp.
TABLE III.— Relative Brightness of Various Papers
IlvIyUMINATED FROM DIFFERENT SOURCES.
Color of paper
White Red Yellow Green Blue
Illuminant $ $ CJ> <p f
Tungsten (Ruby) lamp 100 84 96 23 9
Mercury-vapor lamp 100 8 95 33 14
Tungsten lamp 100 36 91 26 11
Carbon, 4 watts per candle lamp 100 41 94 27 11
The papers were all viewed at the same angle and likewise
illuminated from the same direction. Specular reflection was
carefully avoided. The results for the carbon and tungsten lamps
are the averages of a great many determinations.
SUMMARY.
Based upon purely theoretical considerations, computations are
made showing the influence of colored surroundings on the color
of the useful light. The color of the multi-reflected light is shown
with and without being diluted by direct light. The intensity and
color value of the light is calculated for various reflections. They
were roughly verified by actual measurements of the color
changes of light in a large box, the interior of which was covered
with various combinations of white, yellow, and green papers.
The results obtained were consistent with the previous com-
putations.
The actual change in the reflection coefficients of various col-
ored papers when illuminated by various commercial light sources
is shown.
The writer is indebted to Mr. Leonard Krill for assistance in
the experimental work.
LUCKIESH I INFLUENCE OE COLORED SURROUNDINGS "Jl
DISCUSSION.
Dr. Herbert E. Ives (communicated) : The paper by Mr.
Luckiesh on "The Influence of Colored Surroundings" is not, I
think, as clear as it should be on several points in color measure-
ment.
The first part of the paper contains what are called "purely
theoretical computations" on color changes to be expected by
multiple reflection of light from colored surfaces. As a basis
for these are taken certain "colorimeter readings." As a matter
of fact the computations are not for a colorimeter at all, but for a
spectrophotometer making measurements at three wave-lengths.
Instead of calling these "colorimeter readings" they should be
called "intensities on an arbitrary scale."
Taking up this matter in detail one finds on the second page
that a special illuminant is assumed, radiating three mono-
chromatic radiations — red, green and blue. On the third page
the definition of the quantities NRQ'R, etc., which follows
the calculations shows that the "color values" are supposed to be
obtained by the colorimeter under these three monochromatic
lights. Another limitation which is tacitly made is that the
special tri-chromatic light shall be composed of exactly the wave-
lengths used in the colorimeter.
The complete assumption is then for a tri-color illuminant of
the same constituents as the colorimeter primaries, and that the
colorimeter values of the colored surfaces are to be obtained
under this illuminant. This means nothing more nor less than
that the colorimeter has been ingeniously transformed into a
photometer for making measurements under the three special
colors of light — in other words, into a form of spectrophotometer.
The essential characteristic of a colorimeter, as commonly un-
derstood, is that it measures colors of any type of spectral com-
position in terms of the mixing proportions of three primaries, not
in terms of the relative intensities of the three primary wave-
lengths, as present in the light. A colorimeter will measure a
pure yellow as so many red and so many green, where a spectro-
photometer or Mr. Luckiesh's assumed instrument would find no
red and no green. This illustration will show the caution which
must be observed against confusing the three-color measurements,
72 TRANSACTIONS I. D. S. — PART II
on which calculations have been made in this paper, with colori-
meter measurements as ordinarily carried out. One special
limited case has been chosen in which the colorimeter can be used
as a spectrophotometer. The calculations show the effect of
multiple reflections on monochromatic light. A three-fold repeti-
tion of the expression representing this effect with the introduc-
tion of the subscripts R, G and B does not make it apply to a
colorimeter. The word "colorimeter" had much better have
been left out of this part of the paper.
The problem which Mr. Luckiesh has studied can be solved by
first working out the effect through the whole spectrum (using
many more than three wave-lengths), plotting the results as
spectrophotometric curves and then determining the colorimeter
values either through the color sensation curves, or by the color
mixture curves of the colorimeter used. This process has been
published in some detail in the Transactions of the Society.
The second part stands alone as a set of measurements made
with a different instrument than that "theoretically" discussed and
only related to the first part in a very general manner indeed.
They are "consistent with the theoretical computations" only in
that qualitative way which they must be from the most elementary
considerations of the phenomena of light absorption.
A criticism of the second part is that the manner of use of the
colorimeter is not described. The essential fact to know about
a color mixing instrument is what its primary mixing colors are.
In the Ives colorimeter it makes a difference whether the
primaries are derived from daylight or from a tungsten lamp, or
other artificial source. The description of the instrument as
used should, therefore, be included in a revision of the paper.
A further criticism may be made on the manner of plotting the
colorimeter results. They are shown in a color triangle whose
"white" is the yellow of the tungsten lamp. This means a
triangle not heretofore employed, and one in which the relation-
ships are much distorted from those in the usual white light
triangle. It is the object of scientific study to reduce the number
of elementary quantities to the lowest possible, and to establish
exact relationships between them. The employment of this new
form of triangle without establishing the relationship of its read-
LUCKIESH : INFLUENCE OF COLORED SURROUNDINGS /$
ings to that of a white light triangle is to be regretted. It is one
thing to use different units when the law of relationship is known ;
it is quite a different thing when the law is not known. The
relative positions of points on a straight line may be given with-
out stating the units of measurements, because the relation re-
mains unchanged no matter what the units. In going from one
color triangle to another with different vertices or center, the
units and so the relative positions of colors suffer distortions not
represented by any simple law. It should therefore not be over-
looked that measurements plotted in this triangle are not directly
comparable with those plotted in the usual forms of color triangle.
The measurements are, as they stand much more qualitative than
quantitative.
Mr. M. Luckiesh (in reply) : I am thankful to Dr. Ives for
his interest in the paper.
The theoretical computations made in the first part of the
paper are not for a colorimeter at all nor do I make such a state-
ment. The quantities NRQ'R, etc. , on the third page are not meant
to represent any colorimeter measurements of mine. They do
represent the quantity usually measured in determining the color
of paper. I am well aware of what the computations represent,
for the method was adopted after serious consideration. A spe-
cial illuminant was assumed because of the ease in illustrating
just what I desired to show which was not only the color changes
but also the influence of multiple reflections. I fully appreciate
there are other methods for showing these effects, but I chose
this one as being the most desirable from my viewpoint.
Dr. Ives' discussion contains nothing which attacks the correct-
ness of my computations. No claims for the results are made
knowingly which are not justified by the assumptions.
For the most part of Dr. Ives' discussion which is valuable and
scholarly, consists of considerations I had made before adopting
the method but which I deemed as unnecessary to embody in the
paper.
The latter part of Dr. Ives' discussion takes up the actual
measurements with the colorimeter. I am not in a position as
yet to make my colorimeter measurements accord with those of
Dr. Ives. Therefore it is well to use an altogether different
74 TRANSACTIONS I. £. S. — PART II
triangle. It is true the measurements plotted on this triangle do
not have quite the same spacial relation to each other as they
would have on the triangle used by Ives. But the relative posi-
tions of the color values of tungsten and carbon incandescent
lamps on my triangle are sufficient to give an idea of the order of
magnitude of the changes possible due to colored surroundings.
If the other points are shifted more in one triangle than in the
other all that is necessary in one case is to describe the green
paper as a greener green or the yellow paper as a yellower yellow.
It seems like a waste of time to quibble over that point consider-
ing the difficulties of accurately describing the papers used.
The methods adopted in this paper are legitimate as long as it
is plain that the conclusions are limited by the assumptions. I
intended that they should be so and believe they are. There are
more ways than one to attack this problem but I chose a method
which was clear cut and absolutely free from speculation and am
satisfied that it best illustrated what I desired to bring' forth.
THOMAS : THEORY OF MERCURY-VAPOR APPARATUS 75
THEORY OF MERCURY- VAPOR APPARATUS.*
BY PERCY H. THOMAS.
Synopsis: This paper gives a working hypothesis or conception of
the actions going on in a mercury-vapor apparatus, from the point of
view of the electron theory of electricity. According to the hypothesis
of the paper, current consists of electrons passing in the circuit of the
apparatus and through the vacuum from the cathode to the anode in the
form of material particles. The resistance in the vacuum is due to the
obstruction of the molecules of the gas or vapor in the space, and the
pressure of the vapor is the ordinary pressure of saturated vapor in the
presence of its liquid, as the mercury electrode, and is determined solely
by the temperature of the mercury. The paper also explains the funda-
mental characteristics of the high pressure mercury-vapor lamps in quartz
tubes.
The theory of the mode of operation of the well known mer-
cury-vapor apparatus, characterized by a hermetically sealed con-
tainer exhausted to a high degree of purity and enclosing suitable
positive and negative electrodes, is, more than that of most
physical apparatus, dependent upon the aid of the electron
hypothesis of the nature of electricity. The characteristic ob-
served properties of this apparatus, as distinguished from the
theory of its operation, are very simple in the fundamental em-
bodiment and are well known.
As to the voltage consumed in a mercury-vapor device, it may
be stated that the drop of potential is practically constant for
values of current above a certain minimum, regardless of the
strength of the current, provided the vapor pressure of the
mercury vapor is maintained constant. Of this constant voltage
consumed in the device, there is a certain portion, also constant,
consumed at each electrode, while the remaining constant portion
is consumed in the passage of current through the vacuum or
vapor space. Only this latter portion of the total voltage drop is
dependent upon the length and diameter of the vapor path.
The above statement as to voltage drop applies to the operating
lamp. Before starting, the conditions .are, however, entirely
different. • The device operates as though it contained somewhere
a rigid obstruction to the flow of current, which obstruction sub-
* A paper read at a meeting of the New England section of the Illuminating Engi-
neering Society. Boston, February 17, 1913.
j6 TRANSACTIONS I. E. S. — PART II
stantially disappears when once overcome. A study of the be-
havior of this apparatus, particularly its behavior when utilized
as a rectifier, points conclusively to the surface of the electrode
impressed with potential in the negative direction, that is, to the
cathode, as the location of this peculiar obstruction to the starting.
This fact is established partly by the observed condition that
the starting obstruction, or "reluctance," as it is called by Dr.
Hewitt, can be overcome by various operations at the surface of
the cathode and can not be overcome by any operations in any
other part of the device, and is partly demonstrated by the fact
that, with the device connected according to the ordinary method
as a rectifier, namely, with two anodes connected to the terminals
of the supply and the cathode connected to an intermediate point
of the supply, current will flow freely in any direction in the
vacuum space and will flow out of any electrode impressed with a
positive potential and yet current will not flow between the two
anodes, one or the other of which is always impressed with a
positive potential while the other has a negative potential. Since
now it is known that the anode impressed with the positive po-
tential will not oppose the flow of current from the anode into
the vapor and since it is known that current can flow in any
direction through the vacuum space, it must follow that the
reason the rectifier does not short circuit or "arc" between the
anodes, in other words the reason that current does not flow
directly between the main anodes, is the existence of some sort
of obstruction or reluctance at the anode impressed with the
negative potential, which is in fact the case.
One other characteristic property of the apparatus I wish to
bring out, namely, that the voltage consumed in the vapor path
proper depends upon the vapor pressure or density of the mer-
cury vapor inside the container. Since there is liquid mercury
in the enclosed space and no gas or vapor except the vapor of
mercury, the pressure of the mercury vapor within must always
be the pressure corresponding to the vapor tension of mercury at
the temperature of the liquid mercury itself. This is seen to be
true since, were the vapor pressure less than the appropriate
value, the liquid mercury would evaporate until the pressure of
saturation be reached and were the pressure in the vapor greater,
vapor would be condensed until again the pressure of saturation
THOMAS: THEORY OF MERCURY-VAPOR APPARATUS J J
corresponding to the temperature of the electrode be obtained.
Therefore, the only way to increase or decrease the pressure in
a mercury-vapor tube is to increase or decrease the temperature
of the liquid mercury.
One more point. When mercury evaporates it absorbs heat ;
when it condenses it liberates heat, as in the case of any liquid
in the pressure of its vapor; consequently, if heat be generated
in liquid mercury within the condenser, an equivalent amount of
heat will be transferred to the coolest part of the wall of the
container by the evaporization of mercury at the electrode and
the condensation of mercury on the wall of the container. From
this it follows that where two bodies of mercury, as for example
two mercury electrodes, exist in the same device, they must
necessarily have their surfaces at approximately the same tem-
perature, since otherwise mercury would evaporate from the hot
electrode, thus cooling it, and condense on the cold electrode,
thus heating it. Many of the features and characteristics of
mercury-vapor apparatus can be explained or understood by a
knowledge of these principles. For example, the use of a con-
densing chamber as a means of controlling the temperature of
the lamp that is, that of the mercury cathode, the vapor pressure
and the voltage consumed in the vapor path, may be clearly
understood.
It is now in order to consider the electrical action of the
operation of the device, which for the present discussion may be
assumed to be a lamp. A flow of electricity, according to the
electron hypothesis, to which I personally subscribe, is nothing
more than the passage of electrons along a circuit. It is known
that these electrons are physical bodies of extremely small mass,
requiring between one and two thousand to give the mass of
the hydrogen atom, and carry a definite negative electrical charge.
These electrons exist in a relatively quiescent state in all matter.
Since the charge carried by the electrons is negative and since
unelectrified bodies show no resultant electric charge, it must
follow that in such unelectrified matter- the negative charge on
the electron is balanced by some positive charge in the material.
There will be no flow of electricity in an electric
circuit, that is, no flow of electrons, until something disturbs the
condition of electrical balance of the unelectrified matter. When-
78 TRANSACTIONS I. E. S. — PART II
ever, however, the electron is separated from the positive charge
which ordinarily neutralizes its external electrical effect and
when there exists at the same time an electro-motive-force in the
neighborhood of this electron, the electron will endeavor to fol-
low the electro-motive-force and produce a flow of current. But
electrons like other material particles or bodies can not move un-
less they have a free space in which to move. In all metal con-
ductors it appears that there are passages or spaces between the
atoms or molecules through which the electrons can pass re-
latively freely, although they will experience some resistance (the
well known ohmic resistance of the metals). Electrons set free
in insulating material, however, do not find any passageways
open for motion from one place to another, hence the electrons
with their charges remain fixed in location. This is illustrated,
for example, by the rod of sealing wax and the catskin. When
rubbed by the skin the rod becomes electrified, but the electricity
remains fixed in position. Under this condition the electrons are
stuck, so to speak, in the insulating material.
Although air is ordinarily an insulator it is generally assumed
that air and other gases may be made under certain conditions
to be good conductors of electricity and while, in a certain sense,
that is true, it is, nevertheless, not analytically true. The prin-
cipal reason that electricity does not flow through ordinary air
and gases is that there are no free electrons in the gas to move in
response to an electro-motive-force. When free electrons are"
introduced or produced in a gas, they do pass along through the
gas if a suitable electro-motive-force is present. They pass, how-
ever, not through the atoms themselves but in the space between
them. This action is very much complicated, however, by the
fact that part of the progress may be due to the movement of
the molecules of the air itself and by the fact that electrons have
an attraction for gas molecules and stick to them, sometimes col-
lecting quite a group of molecules called an aggregate. The trans-
fer of electricity in the form of electrons, freed or liberated in
air, is illustrated by the progress of a thunder cloud where the
electric charges travel considerable distances with the air and
probably to some extent through the air itself, slipping between
the molecules. If electrons be set free from the atoms of the
THOMAS: THEORY OF MERCURY-VAPOR APPARATUS /9
gas, as can be done with the aid of X-rays, and an electro-
motive-force be applied, as was done by J. J. Thomson in his
famous experiments in which he applied electro-motive-forces
of opposite signs to parallel plates, there will be an actual flow
of current through the air due to the passage of electrons through
the air. Such currents are always very small, however, since up to
the present time no method has been devised for producing large
quantities of free electrons from gases, under any such circum-
stances.
In the electric arc in air a large number of electrons are
liberated from the cathode and force a lane or passageway to the
anode, probably by crowding back the molecules of air.
The electron and its neutralysing positive charge, when in a
state of equilibrium and constituting a state of non-electrification,
have an attraction for each other and can not be separated with-
out the exertion of a considerable force and the expenditure
of a certain amount of energy and ionization, as such sepa-
ration is called, is ordinarily a difficult process. The forces
being inter- or infra-automatic are very large in propor-
tion to the physical size of the electrons. Furthermore,
it is difficult to employ the powerful forces available in
connection with large masses of matter in such a way as to be
effective in separating the electrons from an atom. Electro-
magnetic waves of very short wave-lengths seem to be effective
in producing this result. Ionization may be produced by such
waves as are supplied by X-ray apparatus or by ultra-violet light.
Another very effective method of producing ionization, that is
the separation of electrons from atoms, is the shock caused by
the striking of one atom by another atom or by an electron pro-
ceeding at a very high velocity.
Electrons are always free and able to move within metallic con-
ductors as long as they do not go outside.
Now the electric circuit containing a vapor electric device
having a high vacuum between two electrodes may be considered.
If an electro-motive-force be impressed in such a circuit the elect-
rons will flow freely in the metal parts of the circuit
from a point of low potential to the point of high
potential (they move backward on account of their carry-
8o TRANSACTIONS I. £. S. — PART II
ing a negative charge) until they reach some point where
their progress is blocked that is where the electrons are not free
to move. They are free to move, however, in all that part of
the circuit constituted by metal conductors. Following the
electrons within these metal conductors one finds them flowing
freely toward the cathode of the device until they reach the sur-
face of the cathode exposed in the vacuum. But the electron can
not leave the surface of the cathode without overcoming the at-
traction of this electron for the corresponding positive charge
associated with it. It could move freely in the body of the metal
since in leaving one positive charge, it could pick up another
from an adjacent atom. These electrons then accumulate at
the cathode surface producing there a negative charge; similarly
electrons are withdrawn from the anode producing there a posi-
tive charge, which two charges impress the electro-motive- force
of the circuit on the vapor path in the vacuum space. Now were
there a supply of free electron in the vacuum space, they would
be immediately drawn out of the vacuum space into the anode
producing a flow of current as long as the supply lasted. As has
already been pointed out there is no material opposition or resist-
ance to the entering of a metal conductor by a free electron since
no counter attraction for a positive charge must then be over-
come. If, however, the electrons which have accumulated at the
surface of the cathode could overcome the attraction they have
for their positive charges and get into the vacuum space they
would leave the cathode and there would immediately be a stream
of electrons between the electrodes in the vacuum space giving a
flow of current and there would be no limit to the amount of
this current unless a limit developed in the supply of electrons.
But the supply of electrons is unlimited in the metals.
Now when a mercury-vapor lamp is started into operation,
this means merely that a means has been provided for liberating
electrons from the surface of the cathode. Then these electrons
are free to flow under the influence of the electro-motive-force of
the circuit from the cathode through the vapor path to the anode
and into the anode through the metallic circuit outside, back
through the cathode lead to the cathode again. In this circuit
there is developed, of course, a certain amount of resistance in
THOMAS: THEORY OF MERCURY- VAPOR APPARATUS 8l
each part of the circuit; the well known ohmic resistance in the
metallic conductors ; a certain resistance to the liberation of
electrons at the cathode surface, and a certain resistance to the
passage of electrons through the vapor space. This latter resist-
ance, namely the resistance to the passage of electrons (or cur-
rent) through the vapor space results principally from the
jostling and blocking of the electrons by the atoms or molecules
of the vapor present which get in the path of the electrons.
The lower the pressure of the vapor the less this resistance and
the less the voltage absorbed in the vapor path; the higher the
vapor pressure, the greater the tendency to impede the progress
of the electrons and the resistance of the lamp or its voltage drop.
In a lamp, however, it is this jostling of the vapor molecules
that produces the light and the more vigorously they are jostled,
that is, the greater the volume of the current flow and the greater
the number of electrons, the greater the amount of light ; and
again the greater the number of vapor molecules, that is the
greater the vapor pressure, the greater the amount of light.
After starting up a mercury-vapor lamp cold, although there
is at first an abnormally large current, there is very little light
produced. This is because there is very little vapor present and
a relatively small number of vapor molecules are jostled. As
the lamp warm? up, however, although the current becomes
somewhat less, the amount of light given is far greater since the
number of molecules of vapor is greatly increased on account
of the higher vapor pressure.
The warming up process comes to a stage of equilibrium when
the heat radiated or dissipated from the surface of the lamp
equals the heat generated in the lamp. If now the heat dissi-
pating capacity (for example by the use of a condensing cham-
ber) of the device is so proportioned that this equilibrium is
reached when the mercury temperature is somewhat above the
boiling point of water one has a mercury-vapor lamp of the low
pressure type; if on the other hand the heat dissipating power
of the lamp is reduced so that equilibrium is reached at a con-
siderably higher temperature, there is obtained the high pressure
type of mercury-vapor lamp, the type for which a quartz con-
tainer may be advantageously used.
82 TRANSACTIONS I. E. S. — PART II
Returning now to the surface of the negative electrode and
the means by which the starting reluctance of the cathode is
overcome and electrons are freed from the body of the cathode
material, it is necessary to confess that the exact mechanism of
this process is not known with certainty. A prominent char-
acteristic feature of the process is, however, the so-called cathode
spot or bright spot of light at the point where the electrons leave
the cathode surface, which is one of the features distinguishing
this light from the so-called Geisler tubes. At this spot some-
thing is going on which is liberating electrons from their asso-
ciated positive charges in the atoms of the liquid mercury. It
may be that the heat generated by the current flow concentrated
at this point produces a very dense vapor and that the current
which is greatly concentrated at this point serves to ionize this
concentrated mercury vapor very energetically, this liberating
of electrons serving to secure the continued flow of current in
the vapor space. In any event there is some result of the flow
of current at any one instant which provides for the liberation
of electrons to constitute the flow of current during the next
instant. Whether this be an extremely local heat effect or the
rapid ionizing of vapor generated locally or whether it be the
liberation of electrons directly from the liquid mercury by the
bombardment of other electrons or positively charged atoms has
not been determined. It is interesting to remember, however,
that if there is an extremely plentiful ionization of vapor at the
cathode spot there will be produced by the current flow first
electrons which will be attracted to the anode and second there
will be liberated by these electrons the corresponding positively
charged atoms, which will be attracted to the cathode by its
negative charge. It may be that these latter atoms which must
be continuously bombarding the cathode are the means of liber-
ating electrons from the cathode to support the flow of current.
However this may be, the fact can hardly be controverted that
the essential action which eliminates the initial starting reluctance
is closely related to some mechanism operating in the cathode
spot and self perpetuating when once started, as long as a
flow of current in sufficient volume continues.
It is a well known fact that if an attempt is made to start a
flow of electricity through an extremely highly exhausted space
THOMAS: THEORY OF MERCURY-VAPOR APPARATUS 83
that enormous potentials are required. It was customary orig-
inally to attribute this phenomenon to the supposed absence of
a conductor in the vacuum space, but our present electron
hypothesis has shown that electricity, that is electrons, being
physical bodies move with the greatest facility in a vacuum and
that the reason that the vacuum device resists the initial flow of
current so stubbornly is the fact that no means exist for lib-
erating electrons within the vacuum space from the surface of
the cathode which is the only point at which they can be pro-
duced, since there is no gas or vapor in the vacuum which can be
ionized to produce electrons. The high voltage required for start-
ing in the high vacuum is to be expected since it is only by forces
acting directly on the atoms themselves that electrons can be pro-
duced from solid or liquid materials and a very high starting
voltage must be provided since it must be applied at a distance.
When, however, the vacuum in the device just discussed is
not perfect and a certain residual gas is present, the high voltage
applied to the terminals is sufficient to ionize this gas producing
electrons and positively charged atoms. These positively charged
atoms, as already described, will be attracted to the cathode sur-
face where they will bombard the material of the cathode, thus
liberating further electrons and positive charges which repeat
the process until under favorable conditions the permanent con-
dition of current flow as already described is attained. The
nature of these phenomena explains why the salient starting
characteristics of the cathode in a vacuum are not observed in
electrodes in the open air, at any rate to anything like the same
extent as in the vacuum. The presence of the air between
electrodes provides a source of electrons and positive charges
automatically sufficient to liberate electrons from the cathode,
whenever a suitable voltage is applied. Furthermore, the pres-
ence of the molecules of air in the path of the current when once
started so greatly increase the operating voltage that the effect
of the starting reluctance would be practically overshadowed.
With this exposition of the hypothesis or conception of the
nature of the operation of a mercury-vapor device which has
satisfied me personally and seems consistent with practically all
the fundamental principles now established in electro physics,
as far as I know them, there remains very little to be said in
3
84 TRANSACTIONS I. E. S. — PART II
explanation of the theory of operation of the practical mercury-
vapor lamp, either the low pressure lamp or the high pressure
lamp in the quartz container.
It may be well to call attention, however, to one characteristic
of the quartz burner of great practical importance in its opera-
tion, though purely incidental in the electrical hypothesis involv-
ing its principle of operation. I refer to the fact that the quartz
burner operating as intended in the Cooper Hewitt commercial
quartz lamp, is approximately a constant current device. That
is if the voltage applied in such a lamp is raised the only effect
is to increase the voltage on the tube without material change
of current through the tube. Of course, the first momentary
effect of the increase in voltage is an increase in the current,
but this increase in current raises the temperature of the lamp
thus increasing the temperature of the mercury electrode and
the pressure of the mercury vapor. This, in turn, increases
the resistance of the lamp or the voltage consumed therein and
the point of equilibrium is found finally at a current only slightly
greater than the original current flow. Lowering of the voltage
merely produces a lowered voltage on the tube when equilibrium
is finally attained with but a slightly decreased current. These
same characteristics are found in the low pressure lamp but are
there not as marked. The particular significance of this phe-
nomenon lies in the fact that if an attempt is made to adjust
the series resistance of a high pressure lamp which is being
installed to the proper value by the insertion of an ammeter in
the circuit, it will be impracticable to make a satisfactory adjust-
ment, since the difference between the current shown on the am-
meter with a very low value of the series resistance and that with
a very high value will be very slight indeed. On the other hand,
if a voltmeter be placed in shunt to the burner it is possible to
adjust the series resistance with great accuracy and certainty,
since the voltage on the burner is very sensitive to the proper
setting. With the use of a voltmeter it is not necessary to pay
any attention to the current, for this will take care of itself.
If it is for any reason desired to increase the current in a
high pressure burner (that is the tube) this must be done by
increasing the natural dissipation of heat from the burner to
give it a lower temperature as by placing it in a cooler place or
THOMAS : THEORY OF MERCURY-VAPOR APPARATUS 85
by directing a draft on the tube or otherwise. In such a case
the net result of the cooling is to lower the temperature of the
electrode and the pressure of the mercury vapor. The result of
the lowering of the voltage or the pressure is an increase in the
current. This increase in the current will then heat up the burner
until the vapor pressure and the burner voltage again bears the
right relation to the supply voltage. Vice versa, when placed in
an abnormally hot atmosphere the burner will take an abnormally
small current. Thus, in very cold weather or with a cracked
globe the tendency of a quartz burner is to take a large amount
of current while its voltage may remain approximately normal.
Another result of this characteristic of the quartz burner is the
difficulty of running the constant potential quartz burners in
series; for suppose a number of such burners to be placed in a
constant current circuit; some of them will naturally run a little
hotter than others either from variations in the structure or from
different temperature conditions at their points of installation.
Those naturally running hotter will be taking a little too much cur-
rent and those running cooler will be taking too little current, but
all are forced to take the same current since they are a constant
current circuit. Since now these devices are constant current
devices, naturally those taking too much current will heat more
and more and the hotter they tend to get (since the added vapor
pressure and resistance from the added temperature increase th&
heat generated in the burner, even if there be no increase of
current). It thus soon happens that a few burners take nearly
all the potential and perhaps ultimately get so hot as to put out
the whole series. The same general difficulty was originally met
with in arc lamps, but was overcome by the use of shunt regu-
lating coils coupled with means for adjusting the length of the
arc between electrodes. The constant potential type of mercury
quartz burner does not provide, however, for such adjustment
and this method is inapplicable.
DISCUSSION.
Mr. C. F. LorEnz (communicated) : It is interesting
to see that the electron, which has come to play such a
familiar role in the every day thought of physicists, is also com-
mencing to penetrate into the transactions of technical societies.
86 TRANSACTIONS I. E. S. PART II
The author presents a very clear and simple picture of the inner
nature of the things going on in a mercury-vapor tube, but in
doing so he throws an even greater burden on the electron than
is customary. He speaks of the flow of current as being entirely
constituted by the motion of electrons. Ordinarily we think of
both positive and negative carriers as entering into nearly every
kind of electric discharge; an exception is the cathode ray stream
in an X-ray tube; another is the discharge of a negatively charged
incandescent solid when located in a very high vacuum. The
author of the paper recognizes the presence of positively charged
particles in the mercury-tube when he speaks of the bombard-
ment of the cathode as the source of ionization at the cathode
during the arc-discharge ; why should such positive charges not
take part in the process of conduction throughout the tube?
Thinking of positive and negative charges leads to an explana-
tion of the "reluctance" which differs from the author's. Before
the arc is started there must be ionization to some extent and the
application of the difference of potential causes motion of the
positive and negative charges which constitutes the minute cur-
rent then flowing; a resistance must immediately develop at both
electrodes owing to the scarcity of ions which immediately exists
in the regions adjacent to the electrodes, since the solid electrodes
do not furnish carriers to replace those swept out of these regions
in the act of conduction. The cathode drop is enormously greater
than the anode drop because of the much greater mobility of the
negative carriers.
A point discussed in the paper of much general interest apart
from its importance in the operation of the quartz mercury
"burner" is the behavior of the latter as an approximately con-
stant current device. Carbon incandescent lamps, Nernst
glowers, metalized filament lamps, and metal filament lamps all
have their own mode of behavior under varying voltage, which
makes them useful as resistances for special purposes, at least in
experimental work. The very peculiarly behaving quartz mer-
cury lamp is a welcome addition to the list of automatically vary-
ing resistors.
Dr. E. Weintraub (communicated) : The electronic theory
of the operation of the mercury arc given by Mr. Thomas does
THEORY OF MERCURY-VAPOR APPARATUS 87
not materially differ from the one current at present among those
working in the field and would, therefore, call for little comment.
I, myself, have on different occasions expressed similar views.
It is true that the electronic picture of the arc is somewhat
vague and indefinite, that especially the role of the positive
electricity is not clear, but these imperfections can be ascribed
and perhaps justly so to the inherent complexity of the arc. From
the point of view of the engineer and inventor a more serious
objection to the theory is the little help it has offered so far in
the technical development of the mercury arc and other arcs. I
for one had to work my own way by direct study of the
phenomena and by the analogy method of reasoning.
However, the electronic theory is the theory of our age, is a
partial expression of the truth, gives a new cross-section through
the infinite complexity of the natural phenomena and whatever
its ultimate fate may be one can hardly do better at present than
to accept it and use it as far as feasible.
It is, however, to be regretted that Mr. Thomas found it neces-
sary to obscure his electronic exposition by still retaining the con-
ception of a "starting reluctance" at the cathode which has no
place in the electronic theory of the arc and for that matter in
any scientific theory. The author himself later explains that the
evacuated space contains no conductive matter, that the latter has
to be produced at the cathode and that an expenditure of energy
is therefore necessary at or near the cathode. If this justifies
the assumption of a starting reluctance then the whole universe
is full of "reluctances." Whenever for lack of energy or for
lack of another sufficient reason a certain transformation does
not happen then there would be an equal reason to postulate a
"reluctance."
With respect to series operation of quartz mercury arc lamps
the evil of running away on constant current which is described
by Mr. Thomas is less pronounced in the case of the new type of
quartz lamp that I have developed recently and what is left of it
is overcome by special automatic regulating means so that at
present the quartz mercury arc lamp is available for both
alternating current and direct current series circuits.
88 TRANSACTIONS I. E. S. — PART II
STREET LIGHTING WITH ORNAMENTAL LUMINOUS
ARC LAMPS*
BY C. A. B. HAIvVORSON, JR.
Synopsis: — This paper outlines the general lighting requirements of
business and residential streets and parkways. For business streets it is
contended (i) that the illumination should be of a different color and
character from that employed in store windows in order that the effect of
the latter may not be impaired, (2) that the illumination should be com-
paratively brilliant, though not greater than the intensity of the windows,
to attract trade and insure traffic safety. Residential street lighting, in the
opinion of the author, requires the use of as few light sources as possible to
produce the average low intensity of illumination, and that a non-uniform
illumination is more desirable than an extremely uniform one. Parkway
and driveway lighting demand primarily an illumination of sufficient in-
tensity from light sources of low intensic brilliancy to insure traffic safety.
In all three classes of lighting the decorative possibilities of ornamental
lamps and standards is said to deserve particular consideration.
In planning a system of exterior illumination to meet the re-
quirements of the city of to-day, the needs of each section
of the city must be carefully considered. The shopping centres
demand one type of illumination peculiarly their own ; the re-
sidential streets, open parks, drives and outlying districts each in
their turn require special consideration and quite different treat-
ment, as regards both the illuminating units and the ornamental
standards or fixtures employed.
Unfortunately, it is quite impossible by the use of illuminating
data alone to show exactly what system of lighting should be
employed in each case to accomplish the best results, as frequently
certain psychological and physiological requirements play parts
of vast importance.
The lighting of business streets requires, in addition to the
usual purposes of good general lighting (police protection, ability
to read easily and distinguish persons, etc.) an illumination toler-
able and pleasing to the eye, yet sufficiently brilliant to produce a
marked effect in the improvement of business — obviously a mat-
ter of great general interest to the city — by attracting people to
the brilliantly lighted thoroughfares and at the same time allow-
* A paper read at a meeting of the New England section of the Illuminating Engi-
neering Society, Boston, February 17, 1913.
halvorson: street lighting 89
ing their attention to be drawn freely to the matter of most im-
portance, which is of course, the attractive window dis-
plays and decorations. This object is largely attained by
means of the contrast between the color schemes employed in
these decorations and the color of the general illumination. It
follows, therefore, that general illumination must be furnished
that differs in color from the illumination obtained from
the small lighting units usually employed in the mer-
chant's local display, and also differing from and not
interfering with the color effects obtained by reflection
or secondary illumination from the goods displayed. White light
falling on the show window from without can only enhance the
color values by showing them correctly. Obviously, then the color
and quality of the light play a most important part in successful
lighting of this kind; accordingly white light of low intrinsic
brilliancy must be employed in order to obtain the best results.
The intensity of the illumination, perhaps, should be con-
sidered of next importance, for the ratio of gen-
eral illumination to window illumination should be such
that the window illumination far outweighs the gen-
eral illumination. As good window display lighting re-
quires in the neighborhood of 15 foot-candles, it is highly im-
probable that any economical scheme of general illumination that
could be obtained, would approach this figure. A possible ex-
ception is yellow flame arc lighting; but this, on account of the
color, would be highly unsuitable, assuming that such illuminants
were placed low, as they usually are in this country, and in the
direct range of vision ; in which case an immense volume of light
would be directed toward the show windows.
Other important considerations are: the appearance of the
lighted unit; the illumination of the building fronts from both
esthetic and economic viewpoints, the latter particularly as re-
gards the benefits accruing the upper-floor tenants; the daytime
appearance of the lighting standard, dignified, simple, or ornate,
as harmony with its surroundings requires, and yet unobtrusive
and free from overhanging arms and glassware that might impede
teaming and endanger pedestrians. Above all the lighting unit
must aid in beautifying the street rather than produce the effect
90 TRANSACTIONS I. E. S. — PART II
of crowding and over-burdening which is so characteristic of
many systems of so-called ornamental lighting.
The problem of laying out a "great white way" system as
described above is quite different from the problem of lighting
the streets of the residential section, for the latter is largely a
utilitarian one and bears on the matter of police protection and
suitable lighting for pedestrians, motorists, and other users of the
thoroughfare.
Residential street lighting is from its very nature of low average
intensity. As this class of lighting comprises a relatively large
percentage of the city's streets, it is important for obvious rea-
sons that the most efficient unit suitable be used. It is desirable,
also, to employ as few light sources as possible for a given
average intensity of illumination, since, the apparent affect
produced by many small light sources is that of a very much
lower intensity of illumination, because the only images pro-
duced on the retina of the eye are those of the light sources
themselves. Especially is this so in the case of oiled roadways
where the amount of light reflected is relatively small.
In residential street lighting, the principle of silhouette light-
ing1 must be employed; that is, seeing is accomplished by the
discernment of objects in contrast with a lighted background,
which usually is the street surface, rather than by light reflected
from the objects themselves.
In the opinion of the writer a non-uniform illumination is more
desirable for this class of lighting than one extremely uniform,
assuming that the minimum intensity in each case is about equal,
but the average intensity higher in the case of the arc lamps than
m the case of low candle-powered units ; and assuming, of course,
the same expenditure of energy per linear foot of street and that
each light source is properly screened by means of a diffusing
globe.
The proper lighting of drives, highways and parkways requires
the use of a specially designed arc lighting unit which
combines white light with maximum efficiency and a low
maintenance cost, as well as a low initial cost when
compared with the installation of many small units of
1 Preston S. Millar, "An Unrecognized Aspect of Street Illumination, Trans., I. E. S.,
p. 546, Vol. V. (Oct. 1910).
HALVORSON I STREET LIGHTING 91
low candle-power. The cost of installation for such light-
ing is a serious factor, for there is usuall) comparatively
little money available for this class of lighting. The prob-
lem of illumination concerns almost wholly the motorist and
drivers of other vehicles. One continually reads of serious night
accidents, involving the automobile usually with a horse-driven
vehicle or a motorcycle, which, statistics show, could have been
avoided had adequate illumination been provided. The require-
ments of such illumination are not greatly different from those
of the residential streets, as described above ; that is, in the
employment of the silhouette principle of lighting. With such
lighting the ability of the eye to see objects clearly a sufficient
distance ahead to avoid collision, is greater than with any other
type of illumination. Obviously, it is necessary that the light
sources should be mounted well above and outside the direct
line of vision, and that they should be of low intrinsic brilliancy,
as too intense a light source destroys the adaptability of the eye
to low average intensity work.
For business streets and other thoroughfares where the re-
quirement is the very best kind of high-intensity illumination, the
great white way lamps* may be used, as they furnish pearl-
while light of low intrinsic brilliancy, produce correct light
reflection from the walls of varied-colored buildings, enhance
the window display, and above all, attract people to the streets
for the improvement of business.
In many cases a satisfactory spacing for units of this type has
* Since the commercial introduction of the ornamental luminous arc lamp at New
Haven, in December, 1911, there have been designed three new forms of ornamental
luminous arc lighting units, known as the "Great White Way" lamp (two forms) con-
suming 520 watts at 6.6 amperes, and 320 watts at 4 amperes; the residential lamp, con-
suming 300 watts at 4 amperes; and the parkway lamp designed for operation at both 4
and 6.6 amperes. 300 and 500 watts respectively. These lamps are shown in Fig. 1.
The great white way lamps are for use on the principal business streets and cities.
The intensity of such illumination of necessity must be high and the distribution good.
Therefore, the units employed should be placed comparatively close together.
The residential lamp, as its name implies, in addition to its use in "Great White
Way" lighting, is also nsed on fine residential streets and on boulevards bordered by
large estates, where as a rule, shade trees overhang the. lamp location. In such cases
the low mounting of the light source (12 feet — 3.66 m.) permits good illumination as it
escape screening by the foliage.
The parkway lamp gives a somewhat more extended light distribution than the two
units just mentioned and is designed especially for roadways where an extremely low
intensity of illumination is adequate for all purposes. It is usually mounted 18 feet
(5.486 m.) above the roadway.
92
TRANSACTIONS I. E. S. — PART II
been found to be approximately 75 to 100 feet (22.86 to 30.48 m.)
on centers on each side of the street, with the standards in a
staggered arrangement. It would be impossible to give here any
pig. 1.— Ornamental luminous arc lamps and standards for (A) residental streets,
(B) "great white way," and (C) parkway lighting.
definite statement on the exact arrangement to be followed for
future installations, as obviously each case must be considered
halvorson: street lighting 93
by itself ; the character, height and color of the buildings, as well
as the width of the street, all playing important parts.
For residential streets and broad boulevards where the screen-
ing effect of foliage must be considered and for all other lighting
classified under the heading "second class," the residential lamp
(see A, Fig. i) can be employed to give the best results.. The
spacing of this lamp, like that of the lamp just described, varies
with the local conditions, but 300 feet (91.44 m.) apart in stag-
gered arrangement gives an exceedingly satisfactory illumination.
An installation of this character will show fewer light sources
within the range of vision, beautify the streets to a much greater
extent, and produce a more satisfactory illumination for the same
cost of installation and maintenance than any other lighting unit
available at the present time. The globes used with these lamps
act as secondary sources of pure white light of low intrinsic bril-
liancy; consequently the units themselves are extremely pleasing
to the eye by night as well as by day, fulfilling all esthetic re-
quirements.
For highways, open parks and drives and all other purposes
where an efficient ornamental unit is desired and where there is
but little danger of screening the light source by foliage, the
parkway lamp (see C, Fig. 1) may be employed in order to utilize
to the highest degree the principle of silhouette lighting.
With these three units it is believed that the complex and
exacting requirements of scientific street illumination can be
successfully met, both from a utilitarian and from an artistic
viewpoint. The introduction of the smaller and consequently
lower candle-powered units will be greatly appreciated by those
cities which possess at the present time an installation of the
pendant type of standard luminous arc lamps, as these new orna-
mental units are interchangeable with the pendant type units
insofar as their operating characteristics are concerned, and con-
tain many vital mechanism parts such as magnets, clutches, etc.,
which are common to both. This feature is one which will be
greatly appreciated by the operating man.
94 TRANSACTIONS I. E. S. PART II
DISCUSSION.
P. S. Millar (communicated) : Mr. Halvorson's paper is
somewhat radical in that it departs from the traditional view that
small illuminants should be employed on side streets and in
residence districts.
Two points are particularly worthy of note, although some may
question the correctness of the author's view. The first is that
comparatively brilliant illumination of a thoroughfare does not
detract from window displays if there is considerable contrast
between the color of the street light and that employed in the
windows. The other is the importance of the appearance which
the street lighting fixtures present in the daytime, and the part
which they play in the general appearance of a street. Both
points well merit greater consideration than has been given them
in the past.
The silhouette effect in street lighting, which is emphasized so
strongly in this paper, is of greatest importance where the re-
quirements are for the discernment of large objects. Lighting
of streets with few large units rather than with many small ones
is likely to promote this form of seeing. It must not be for-
gotten, however, that the requirements for the discernment of
small irregularities in street surface are important, particularly
to the pedestrian and the driver of a slow moving vehicle. For
any given street and lighting appropriation there is probably
some superior balance between the requirements for the discern-
ment of large objects by silhouetting and the discernment of
street surface irregularities, and this is likely to imply some
balance between number and size of light sources.
The illuminating effect is one of a number of considerations
which determine the choice of a street lighting system. Insofar
as it is the leading consideration, our understanding of street
lighting requirements is rather vague. Much remains to be done
in the study of street illumination before the principles which
determine this balance between number and size of units can be
announced. Until then, practise must be guided by conclusions
drawn from views of individual experts, one extreme of which
appears to be represented in this paper. It is the logical con-
clusion of this extreme view that in all classes of streets, lamps
STREET LIGHTING 95
which are of relatively large power may be employed to secure
best results. There will probably be many who will take issue
with the author in regard to this conclusion. It is to be hoped
that the study of this question may be undertaken seriously, with
a view of answering some of the moot questions which are sug-
gested by this paper.
Mr. W. A. Darrah (communicated) : I believe that the
writer has given a very clear discussion of his subject, but think
that perhaps some points are of sufficient importance to warrant
a further discussion. It is true that from the standpoint of the
effect produced the illumination of our city streets is a matter
which cannot be fully expressed in scientific formulae because of
the psychological element involved. There are certain points,
however, which are capable of rather exact determination.
Whether the lighting of residential sections is considered or
whether the illumination is confined entirely to that of business
districts, it is my opinion that the light source should be placed
well above the range of vision in all ordinary cases, 25 feet to
35 feet, being a desirable height. In fixing upon the elevation of
the light sources, consideration should be given to the illuminated
signs along the street as well as to the store windows. It is
obviously undersirable to mask the effect of an expensive
illuminated sign by an arc lamp placed closely adjacent to it.
I believe the writer's contention that a light source of relatively
low intrinsic brillancy is desirable, is correct. In this connection
I believe the flame carbon arc is a decided step in advance, particu-
larly when used with diffusing glassware. Owing to the relatively
large source of light the intrinsic brilliancy of the flame arc is
considerably less than that of the older types of arcs.
A further consideration of the subject of street lighting and
one which has not been discussed in the present paper is the sub-
ject of station apparatus. In deciding upon any general system
of lighting, the subject should be considered broadly and in addi-
tion to the specific lamps involved, the generating and accessory
apparatus should be considered. The simplicity of equipment
and low cost of maintenance of alternating current apparatus is
a further fact which tends to emphasize the value of the flame
carbon arc lamp as a unit for street lighting.
96 TRANSACTIONS I. E. S. — PART II
The flexibility of this system in the way of color and in-
tensity is an additional point which should be given full weight
in considering a street lighting system. As an example of this
flexibility may be sighted the use of white light for the general
illumination of business streets with the use of the same lamps
equipped with carbons for producing yellow light at busy street
corners, fire alarm boxes and other points which require special
identification.
J. R. Cravath (communicated) : The position taken by the
author that it is desirable to employ as few light sources as
possible for a given average intensity of illumination on
residential streets: this theory is obviously subject to consider-
able limitation because if it were carried to its logical conclusion
we would have but one source of light for an entire length of
street.
The author's opinion that non-uniform illumination is more
desirable than illumination which is extremely uniform for this
class of lighting is open to considerable question. If carried to
its logical conclusion, this would drive us back to the old open arc
lamp for street lighting, and we would have to discard the later
types of lamps such as the magnetite arc which gives a more uni-
form distribution of light along the street. Within the past few
years considerable intelligent effort has been directed toward pro-
ducing more uniform illumination along a street on the theory
that the extreme contrast between the areas of high and low
illumination is detrimental to seeing clearly, as is also the blinding
effect of high candle-power lamps spaced at infrequent intervals.
There is no doubt whatever that changes in the direction of better
uniformity have resulted in a decidedly improved street lighting
in many residence streets. But there is some question how far
we can go in the direction of greater uniformity to advantage,
and the point raised by Mr. Halvorson is a very interesting one
which should be the subject of thorough investigation before con-
clusions are drawn as to how far it will pay to go in securing
uniformity of light distribution.
Prof. J. M. Bryant (communicated) : The author has ap-
parently made a careful study of the general conditions required
for lighting in a medium sized city. Too many cities have made
STREET LIGHTING 97
the mistake of putting all their available money into the "great
white way" because they have obliged to appease the business
men who are influential citizens paying a considerable part of
the city taxes.
Within the last few years another portion of the city, namely
the better residential portions, parks and boulevards have re-
ceived more attention since this same influential citizen has found
it a pleasure to be out in the evening in his automobile. As yet
the poorer business and manufacturing portions of the city as
well as the poorer residential portions have received but little con-
sideration as to lighting. Since it is here that the greatest amount
of crime exists our aim should be to light and care for these dis-
tricts in a much better manner, in order to improve the character
of the inhabitants and to aid in police protection. We should be
as ready to spend the city's money for this purpose as we are our
own for charitable or police work among these inhabitants.
No one form of lighting unit can be employed with equal
effectiveness or economy in all parts of a city. The arc lamp
due to its high candle-power must be used for lighting in places
requiring strong light and also rather wide distribution. How-
ever, due to the relatively high coefficient of reflection of build-
ings, this type of unit may be used in business streets. The light
reflection from buildings tends to bring up the low intensity
areas and make the illumination uniform if the lamps are not
too widely spaced. In park lighting and broad boulevards the
arc light may be employed to advantage, provided enclosing
globes are used, thus lowering its intrinsic brilliancy.
In lighting the streets in residence quarters the lower candle-
power unit obtained only in incandescent lamps such as the series
tungsten lamp should be employed. On account of shade trees
and other features these lamps must be hung relatively low,
easily coming into the general line of vision. It has been found
advisable to screen even those lamps by globes to make it easier
for the eye to distinuish distant objects. I can not agree with the
author in the use of arc lamps widely spaced for this part of the
installation.
<\<
TRANSACTIONS
OF THE
Illuminating Engineering Society
Published monthly, except during July, August, and September, by the
ILLUMINATING ENGINEERING SOCIETY
General Offices: 29 West Thirty-Ninth Street. New York
Vol. VIII
MARCH. 1913
No. 3
Council Notes.
A regular meeting of the council was
held in the general offices of the society,
jm West 39th Street, New York, March
14, 1913. Those present were : Preston
S. Millar, president; C. O. Bond, Percy
W. Cobb. Joseph D. Israel, general sec-
retary; A. E. Kennelly, V. R. Lansingh,
Norman Macbeth, L,. B. Marks, treas-
urer; C. J. Russell and W. J. Serrill.
An oral report supplemented by a
statement of the society's expenses and
membership covering the first two
months of 1913 was received from the
general secretary.
An estimate of the expenses and in-
come of the society for the fiscal year,
January 1 to October 1, 1913, was re-
ceived from the finance committee. The
income was given as $8,870; the ex-
penses $8,800.
The following appropriations were
authorized: $245 for a typewriter and
dictation outfit; $80 for printing and
$20 for a membership campaign by the
Chicago section.
Payment of vouchers Nos. 1210 to
1240, inclusive, aggregating $790.18 was
approved upon recommendation of the
finance committee.
Progress reports were received from
the committee on collegiate education,
the committee on sustaining member-
ship, the committee on section develop-
ment and the committee in charge of
the organization of the New Lake Erie
section.
In the report of the section develop-
ment committee the desirability of es-
tablishing a student grade of member-
ship was raised. The secretary was
directed to forward to the committee
on collegiate education the information
and data on the subject contained in
the report.
Monthly reports on the activities of
sections were received from the follow-
ing vice-presidents : J. R. Cravath,
Chicago section; J. W. Cowles, New
England section; W. J. Serrill, Phila-
delphia section; Howard S. Evans,
Pittsburgh section.
The illumination primer committee
reported that it had effected a few minor
changes in the primer, also that an
order for 100,000 copies was pending.
A progress report was also received
from the committee on reciprocal rela-
tions with other societies.
In accordance with a recommendation
of the latter committee the president
was directed to appoint committees to
care for the Illuminating Engineering
Society's part in the fourth Interna-
tional Congress for School Hygiene in
Buffalo, August 25 to 30, 1913, and the
convention of the American Gas Insti-
tute at Richmond, Ya.. next fall. The
arrangement of the joint meetings which
TRANSACTIONS I. E. S. — PART I
will be held on the occasion of both
the latter-mentioned events is the result
of activities of the committee on recip-
rocal relations with other societies.
It was decided to have available for
joint meetings with other societies 1,000
copies of the primer for free distribu-
tion. It was understood that these
copies would be marked "Compliments
of the Illuminating Engineering Society"
and have on the inside a notation about
the work of the society and the prices
of the primer.
The committee appointed to devise
plans for the appointment of local rep-
resentatives in cities not having sections
of the society reported that it had con-
sidered "the question of proper pro-
cedure in the selection of local repre-
sentatives and had come to the conclu-
sion that this is a matter in which each
locality presents a special problem by
itself. It was thought, therefore, that
as questions of local representatives
arise they should be handled by the
general officers of the society who
should consult with the members in the
locality in question and use such other
means as may seem best to them to
arrive at a proper conclusion." The
report was accepted and the committee
discharged.
It was resolved that the advertising
committee should be instructed to com-
plete negotiations for advertising to the
extent outlined in dieir annual report
and that further negotiations for ad-
vertising in excess of that reported be
not made.
President Millar reported upon prog-
ress of plans for the reorganization of
the International Photometric Commis-
sion. President Vautier of that Com-
mission has appointed a sub-committee
consisting of representatives of the
several national laboratories, which sub-
committee has been asked to formulate
plans for reorganization of the Com-
mission in such a way as to make it
thoroughly representative of illumina-
tion interests. Such plan is to be sub-
mitted to the several national gas socie-
ties for approval. Dr. E. B. Rosa of the
Bureau of Standards is the American
representative upon the sub-committee.
Proposals of the sub-committee con-
template for this country the issuance
of an invitation by the American Gas
Institute to other national societies in-
terested in light, illumination and pho-
tometry to meet for the formation of a
national committee, which committee is
to delegate representatives to represent
the American committee at the first
meeting of the proposed Commission,
which is projected for Berlin in Sep-
tember, 1913. All of these plans are
tentative and subject to change, and the
ultimate arrangement may be materially
different from that suggested. The
present status of the work of the sub-
committee is, however, as indicated.
The following letter was received
from Mr. Comfort A. Adams, secretary
of the Standards Committee of the
American Institute of Electrical Engi-
neers :
"At the last meeting of the Standards
Committee, it was voted to present to
the Board of Directors of the A. I. E. E.
the following by-law for their adoption :
'The Standards Committee of the
A. I. E. E. is instructed by the Board
of Directors of the A. I. E. E. to take
no action on any subject matter out-
side of the field of electrical and
magnetic standardization and within
the field of the Standards Committee
of another national society, before
coming to an agreement with the
Standards Committee of that society,
provided that the said society instructs
TRANSACTIONS I. E. S. — PART I
its Standards Committee not to take
action in electric or magnetic stand-
ardization before coming to an agree-
ment with the Standards Committee
of the A. I. E. E.'
"As a thirty days' notice is required
before the vote can be taken, the Board
of Directors will not be able to act
upon this until their April meeting. In
order to avoid delay, the Standards
Committee urge that similar action be
taken by your society."
It was resolved that the committee on
nomenclature and standards of the
Illuminating Engineering Society be
urged to adopt a resolution similar to
that contained in Mr. Adam's communi-
cation.
President Millar was authorized to
appoint a committee on time and place
for the 1913 convention.
President Millar announced appoint-
ments to the various standing and tem-
porary committees of the society which
he had made since the January council
meeting. These appointments were
approved.
New Members.
The following applicants were elected
members of the society at a meeting of
the council held March 14. 1913:
Argabrite. H. M.
Manager, Elwood Electric Light
Company, Elwood, Ind.
Bush. W. E.
Illuminating Engineer. The British
Thomson-Houston Company, Ltd.,
J7 Upper Thames Street, London.
Eng.
Hull, Schuyler M.
Circuit Breaker Engineer, Westing-
house Electric & Mfg. Company,
East Pittsburgh, Pa.
Johnson, C. W.
Salesman, Westinghouse Electric &
Mfg. Company, 1205 Dime Savings
Bank Bldg., Detroit, Mich.
Junes, Geo. A.
Salesman. Macbeth-Evans Glass
Company, Pittsburgh, Pa.
Kilmer, W. S.
Lighting Engineer, H. W. Johns-
Manville Company. New York. N. Y.
Kirkpatrick, R. B.
Salesman. The Philadelphia Electric
Company, 1000 Chestnut Street,
Philadelphia, Pa.
Kxauber, Alex. M.
President, Alex. M. Knauber Com-
pany, 742 Euclid Avenue. Oak Park,
111.
Lyxch. James D.
Chief Engineer, Lit Brothers, 6th
& Market Sts., Philadelphia, Pa.
Muxroe, Roy G.
Service Supervisor, The Denver Gas
& Electric Company. Gas & Electric
Bldg.. Denver, Col.
Pearsox, Julius T.
Lamp Salesman. Westinghouse Lamp
Company, 27 Woodward Avenue,
Detroit. Mich.
Ross, Jay H.
Secy.-Treas., American Electrical
Equipment Co., Kansas City, Mo.
Smullex. R. W.
The Philadelphia Electric Company,
1000 Chestnut Street. Philadelphia.
Pa.
Young, A. W.
Manager. New Business. Public
Service Electric Company. 418
Federal Street. Camden, N. T.
Sustaining Members.
The following companies were elected
sustaining members of the society at a
meeting of the council March 14, 1913:
TRANSACTIONS I. E. S. — PART I
Electrical Testing Laboratories.
Holophane Works of General Electric
Company.
The Edison Electric Illuminating
Company of Boston.
The New York Edison Company.
The Philadelphia Electric Company.
Section Activities.
CHICAGO SECTION
At a regular meeting of the Chicago
section in the auditorium of the Western
Society of Engineers, Monadnock Block,
March 12, Mr. Meyer J. Sturm pre-
sented a paper entitled "Practical Ideal-
ism in Illumination with Particular
Reference to Hospitals." The paper
will appear in a later issue of the
Transactions. Eighty members and
guests were present.
NEW ENGLAND SECTION
The New England section held a joint
meeting with the New England section
of the National Commercial Gas Asso-
ciation in the auditorium of the Edison
Illuminating Company Building, 39
Boylston Street, Boston, March 12.
Mr. Preston S. Millar, president of the
I. E. S., presented a paper on "Some
Phases of the Illumination of Inte-
riors" ; the paper was supplemented by
a demonstration of lighting effects in
miniature rooms. The paper and dis-
cussion appears in this issue of the
Transactions.
NEW YORK SECTION
The New York section held a meet-
ing in the United Engineering Societies
Building, 29 West 39th Street, New
York, March 13, 1913. Three other
societies — New York Association for
the Blind, Committee on Prevention of
Blindness; American Society of Me-
chanical Engineers, American Museum
of Safety — participated in the meeting.
Three papers were presented : "Illumina-
tion and Eyestrain" by Dr. Ellice M.
Alger; "Mechanical Safety" by Dr. W.
H. Tolman, and "Industrial Lighting" by
Ward Harrison. The first one of these
papers appears in this issue of the
Transactions.
philadelphia section
The March meeting of the Philadel-
phia section was held on the 28th of
the month at the Pennsylvania Academy
of the Fine Arts. Fifty ladies and
gentlemen were present at the dinner
at the Hotel Walton preceding the
meeting. Through the courtesy of the
directors of the Academy, the art gal-
lery was open for inspection before the
meeting, and this proved a very enjoy-
able feature. At the meeting Mr. M.
Luckiesh of the National Electric Lamp
Association presented a lecture and
demonstration on the subject of "Light
and Art." By means of demonstrations,
Mr. Luckiesh showed the effect of di-
rection, color and quantitative distribu-
tion of lights on objects of architecture,
sculpture and paintings. About 140
members and guests were present.
PITTSBURGH SECTION
A meeting of the Pittsburgh section
was held in the auditorium of the
Engineers' Society of Western Pennsyl-
vania, Oliver Building, March 14, 1913.
Mr. M. Luckiesh of the National Elec-
tric Lamp Association, Cleveland, O.,
presented a paper entitled "Light and
Art" which was supplemented by a
series of demonstrations showing the
effect and influence of quality and direc-
tion of light on various art objects.
Thirty members were present.
TRANSACTIONS
OF THE
Illuminating
Engineering Society
MARCH, 1913
PART II
c^
f%
Papers, Discussions and Reports
[ MARCH, 1913 ]
CONTENTS - PART II
Some Phases of the Illumination of Interiors. By Preston
S. Millar 99
Illumination and Eyestrain. By Ellice M. Alger 130
Home Illumination. By George Leland Hunter 149
11
SOME PHASES OF THE ILLUMINATION OF
INTERIORS.*
BY PRESTON S. MILLAR.
Synopsis: This paper is an exposition of some of the fundamentals
of interior lighting. It treats of the questions of glare, diffusion, direction
of light, contrast, influence of colored surroundings on illumination, etc.,
and describes and illustrates some of the ordinary methods of lighting
now in use. The introduction presents a brief discussion of the work
and functions of the Illuminating Engineering Society and the illuminating
engineer.
I.— THE ILLUMINATING ENGINEERING SOCIETY AND ITS
RELATION TO PUBLIC LIGHTING COMPANIES.
The business of the lighting company is to sell illuminating,
power and heating service. Its aims and achievements are
broader than the mere selling of electrical or gas energy. In
the conduct of its business, the well organized large lighting
company must have on its staff, power experts, heat and ventila-
tion experts, and illumination experts. The illumination expert
should be well qualified to deal with lighting problems, and to
the extent that his services are applied on behalf of the com-
pany and its customers, the lighting company may be said to be
engaged in illuminating engineering work. Such activity, how-
ever, forms but an insignificant part of the company enterprise.
While its importance is more generally recognized than for-
merly, such illuminating engineering of necessity is restricted
very largely to the application of existing knowledge, rather
than to the development of new knowledge in the field of illumi-
nation. The lighting company applies developments, and it is
only natural that the art of illumination (art in the sense of
application of knowledge) should be emphasized in lighting com-
pany illuminating engineering, and that the development of the
* A paper and demonstration given before the following sections of the Illuminating
Engineering Society ; New Vork, November iS, 1912, Philadelphia, February 21, 1913,
New England, March 12, 1913.
IOO TRANSACTIONS I. E. S. — PART II
science of illumination should receive but little impetus therefrom.
Thus associations of lighting companies are interested in the
practise of illuminating engineering. And the discussion of
improvements in lighting practice is a feature of increasing inter-
est in the meetings of such associations. There are, however,
other aspects of illumination not properly included in these pro-
ceedings which may not be neglected if the science and art are
to be developed properly.
The problem of illuminating engineering may be summed up
in a practical way as follows : Illumination must be provided
with a view to rendering visible the things which it is desired
to have seen. The illuminating engineer, in studying each prob-
lem, must ascertain what it is desired to have seen, and for this
purpose, must inform himself concerning the requirements and
the viewpoints of those who are furnishing the lighting, and
those who are expected to see by its aid. In a machine shop, the
illuminating engineer must put himself in the place of the
mechanic, and must so design the installation that the mechanic
can see the surfaces upon which he works and can see to apply
tools properly. In the building in which the architect has sought
for certain effects, the illuminating engineer must provide light
with which to display, in a proper way, those surfaces and orna-
mentations which the architect desires to have seen. Briefly, the
illumination must be designed for the particular purpose for
which it is to be used.
The illuminating engineer must not only render visible the
things which are to be seen; he must also establish and maintain
hygienic conditions for the eyes and body. It must be prac-
ticable to see without injury to the eyes, and without discomfort.
It may be that in the proper discharge of this part of his func-
tions, the illuminating engineer may be called upon to go beyond
his strict province, and to influence conditions other than those
of illumination. For example, he may have to urge the use of
suitable paper in schoolbooks.
The illuminating engineer must not only render visible the
things which are to be seen, and make vision possible under
hygienic conditions ; he must also consult esthetic requirements,
and conform to correct principles of architecture and decora-
tion, thereby satisfying discriminating taste. He must choose
MILLAR: ILLUMINATION OF INTERIORS IOI
fixtures and lighting equipment which will be in harmony with
the character of the installation and the decorations. He must
so distribute, diffuse, and modify the color of the light, as to
produce pleasing effects.
These three requirements having been met — a tasteful and
satisfying installation having been provided, with which it is
possible to see with comfort the things which it is desired to have
seen — the illumination may be said to be effective, and the work
of the illuminating engineer may be said to have attained one
of its primary objects. There remains the important and funda-
mental consideration that these things must be accomplished
with reasonable economy. If large energy consumption must
be incurred in order to make the installation effective, the illumi-
nating engineer does not hesitate, for the installation is efficient
nevertheless. It cannot be efficient unless it achieves the purpose
for which it is designed, and large energy consumption or high
maintenance cost does not necessarily imply inefficiency. But it
is found usually that there is ample opportunity for the expert-
ness of the illuminating engineer to manifest itself in so design-
ing the installation that it shall be effective in accomplishing the
purpose for which it is intended, while reducing the cost very
materially below that which would be required in order to secure
the same lighting effects by inexpert methods.
Here, then, is the illumination art in a nutshell : to render
visible the things which it is desired to have seen; to establish
hygienic conditions for vision; to conform to esthetic require-
ments; and to accomplish these things with reasonable economy.
It is conceivable that in designing a building those responsible
might desire to provide an illuminating equipment which should
be as nearly as possible perfect. Having investigated the prob-
lem, it might be determined to retain a group of specialists in
the various professions which are concerned with the problem.
It is conceivable, then, that there would be retained an engineer
who is conversant with the properties of illuminants ; a con-
tractor who would install the equipment; a fixture designer com-
petent to design or select suitable lighting fixtures ; a glass expert
competent to produce such quality of glassware as it might be
desired to employ in the installation; a physicist qualified to
apply optical laws in the design of reflecting and diffusing sur-
102 TRANSACTIONS I. E. S. — PART II
faces ; an accomplished decorator ; a psychologist and an ophthal-
mologist, who would pass upon the conditions from the viewpoint
of vision conservation. These specialists, in co-operation with
the architect of the building, would meet to decide upon plans for
providing illumination. The viewpoints of each would be impor-
tant to an extent that would make neglect seriously prejudicial
to the success of the lighting installation. Each specialist might
be expected to know, or to think that he knows, what kind of
illumination is necessary in order to fulfill the requirements from
his viewpoint, but he would not know how to obtain such illumi-
nation. Furthermore, each specialist would be more or less
ignorant of the illumination requirements, judged from the
standpoints of the other members of the committee. Whether or
not such a committee of experts would be able to agree upon a
particular plan for illuminating the building, can be left to the
imagination.
The well-qualified illuminating engineer must be informed in
regard to the underlying principles of each science and art which
would be represented by a specialist on our hypothetical com-
mittee. It is too much to expect that he should be an expert in
each of these lines :
"A man so various that he seems to be
Not one, but all mankind's epitome"
He must, however, have sufficient knowledge of the fundamentals
and sympathy with the aims of each of the sciences and arts in-
volved, to bring his work into harmony with them, and it may be
noted that he should be better qualified than would the committee,
to produce a well-balanced design, because his knowledge of the
other phases should enable him to give each its proper weight in
final consideration of the subject.
To state the function of the illuminating engineer, is to indicate
the work of the Illuminating Engineering Society. It seeks to
be the forum where specialists, engaged in each of the sciences
and arts which enter into illuminating work, can meet for study
and discussion of the problem, exchanging views, learning from
one another, and endeavoring to establish correct principles upon
which illuminating engineering must be based. Once established,
through the meetings and Transactions of the Society these
millar: illumination of interiors 103
principles are quickly disseminated among the membership, and
are applied in the practical work of illuminating engineers. When
proven beyond peradventure, effort is made to make them known
to the public, as in the case of the Illumination Primer, recently
published by the Illuminating Engineering Society.
This brief discussion should serve to indicate the character of
the Illuminating Engineering Society. There is a real need for
an organization which shall make a specialty of illumination,
developing the science and the art so that it may be applied by
lighting companies and others. I believe that an impartial study
of illumination developments of the past six years will lead to the
conclusion that the Illuminating Engineering Society has justified
itself by its achievements and by its present status.
II.— DEMONSTRATION OF LIGHTING EFFECTS.
It is a peculiarity of illuminating engineering that the demands
are for the highest technical knowledge and skill applied in the
common walks of life ; that success or failure affects closely the
people in their ordinary occupations. The most technical dis-
cussion of subjects pertaining to illumination is likely to have a
practical application, of interest to the man in the street. It is
the purpose of this presentation1 to indicate in a non-technical
manner some of the features of the illumination of interiors
which have been studied, and upon which we have some informa-
tion as a basis for practise.
If a room be illuminated by a bare lamp (Fig. 1, right), the
results are unsatisfactory for a number of reasons. In the first
place, the walls receive the major amount of the light produced
and the portions of the room in which the light is more likely to
1 For the lecture upon which these notes are based miniature rooms were constructed.
These were 4 by 4 feet and 3^ feet high. Wall decorations, as well as lighting equip-
ments, could be altered readily. With one or two exceptions lamps were so operated as
to produce 64 lumens in each room. This permitted of comparison of various lighting
systems on an equitable basis. With this demonstration equipment a wide variety of
lighting effects could be produced and approximately forty were projected. Time limita-
tion, however, restricted those actually presented to twelve which are here illustrated by
photographs available through the courtesy of The New York Edison Company.
The temptations to deal with vaiions types of interiors and to undertake studies of
color were resisted. Only one type of interior, and that in its simplest form, was consid-
ered, time lacking for a more extended discussion.
The suite of miniature rooms was constructed under the supervision of Mr. W. F.
Little and was operated under his direction by Messrs. H. Bardwell, M. D. Beuick and W
Ihlefeld, all of the Electrical Testing Laboratories.
104 TRANSACTIONS I. K. S. — PART II
be utilized, are inadequately illuminated. The light source is un-
attractive, and, when within the field of vision, is annoying, if
not actually injurious to the eyesight. This latter effect, included
under the name of glare, is very noticeable in the illustration,
where the lamp is at the center of the field of vision, and the effect
is exaggerated beyond that which would be experienced by oc-
cupants of the room.
If the lamp be shielded from view (Fig. I, left), conditions are
much improved. Much of the discomfort and annoyance dis-
appears. While the distribution of light on surfaces seen within
the room is not changed materially, yet everything can be seen
more distinctly. Observe, for example, the vertical stripes upon
the wall-paper. Beginning near the floor, trace a vertical stripe
which is at some height, almost, if not quite in line with the lamp
in room No. i (to the right). It will be noticed that as the gaze
approaches the vicinity of the lamp, it is difficult to see the stripe,
and that when level with the lamp, the stripe disappears entirely.
In room No. 3 (to the left) in which the lamp is shielded from
the eye, the corresponding stripe may be distinguished when look-
ing just past the lamp and screen.
One of the important functions of a reflector or other lighting
auxiliary is to thus shield the lamp from view, by interposing be-
tween it and the eye, either an opaque or a translucent medium.
This is accomplished in room No. 2 (Fig. 1, middle).
But a reflector should fulfill other, equally useful, purposes.
In shielding the lamp from view, it may also be made to direct a
considerable proportion of the light where it can be utilized to
best advantage. Much study has been given to this aspect of the
problem, and the performance of any standard type of reflector
may be ascertained by reference to photometric tests of light dis-
tributed downward may be judged from the floor brightness,
day prepared to supply with their wares. Perhaps in no branch
of illumination have such great strides been made in the past ten
years as in the design of reflectors in particular, and lighting
auxiliaries in general.
It has been noted that the bare lamp distributes but a small pro-
portion of the light downward. In room No. 2 (Fig. 2) a
reflector is employed which re-directs downward a goodly propor-
MILLAR: ILLUMINATION OF INTERIORS 105
tion of the light, illuminating the card below it much more
brightly than would a bare lamp. In room Xo. 3, this re-direction
of light is effected in such a way as to concentrate a large propor-
tion directly below the lamp, thereby illuminating the card to a
brightness which is about twice that of the card in room No. 2,
which was considerably brighter than the card in the room where
no reflector was used.
In room No. 1, as now equipped, a reflector is employed which
has been designed without regard to optical laws, and which,
though looking like a prismatic reflector, has in fact almost none
of the qualities which characterize such glassware. It accom-
plishes little in the way of re-direction of light, while affording
but an ineffectual protection for the eyes against brightness of the
filament. It absorbs a certain amount of light without rendering
any adequate return in improvement of conditions.
In the three rooms there is illustrated the range of practicable
accomplishment in the employment of reflectors, if we omit
opaque reflectors, which would not be suitable for employment
under such conditions. In room No. 1, general distribution of
the light throughout the room ; in room No. 2, effective re-direc-
tion of much of the light downward, largely increasing the
intensity on the table plane, though illuminating the walls and
ceiling brightly enough to avoid the appearance of dimness. In
room No. 3, the concentration within a small area beneath the
lamp is very marked, this being effected by taking from the
walls and from the table plane near the walls, a portion of the
light which falls upon them in room No. 2, and concentrating
it upon, or near the table. The relative intensities of light dis-
tributed downward may be judged from the floor brightness.
The correct design of a reflector to accomplish a given purpose,
involves the application of well-known optical laws. With pris-
matic glass and mirror types of reflectors, a wide variety of
distribution may be obtained. With opal or phosphate glasses,
such as that in room No. 2, the possibilities of securing high
concentration are rather more limited, though with this one
exception these too may be designed to produce practically any
distribution likely to be required.
In achieving the particular distribution which characterizes a
given reflector, it is important that the light source be correctly
106 TRANSACTIONS I. E. S. — PART II
located with reference to the reflector. The use of an incorrect
shade holder, or of an improper lamp distorts the distribution
and usually detracts from the appearance and usefulness of the
lighting unit.
In reflectors, as well as in globes and other forms of glass
lighting auxiliaries, the degree of optical density is important,
affecting both the performance and the appearance of the glass.
This is an important feature to be considered in selecting glass-
ware. In the now rather common forms of display street light-
ing, which utilize clusters of tungsten lamps in globes, very dis-
pleasing effects are sometimes encountered, due, first, to the non-
uniformity of the globes, and second, to the insufficient density
which makes the location of the lamp apparent, instead of
rendering the whole surface of the globe equally bright, making
it appear a ball of light. Much of the lighting glassware in use
in residences a few years ago, and it is to be feared even to-day,
consists of etched or frosted crystal glass which serves chiefly
to give the fixture a somewhat finished appearance. It neither
directs sufficient light usefully to make it efficient, nor conceals
the light source sufficiently to make it attractive or of value in
protecting the eyes.
In Fig. 3 there is a globe of crystal glass, roughed inside, a
light opal globe, and a denser opal globe. The last presents a
better appearance without involving serious sacrifices otherwise.
The light absorptions of these balls are respectively :
Frosted ball 6%
Light opal ball 13%
Dense opal ball 22 %
\Yhen employed in the miniature rooms shown in Fig. 2 the
relative light intensities throughout the table plane averaged:
Frosted ball 100%
Light opal ball 106%
Dense opal ball 95 %
In passing I wish to refer to the great diversity of lighting
fixtures and glassware which are now available in standard
types, awaiting selection. Practically all ordinary requirements
in interior illumination may be filled by lighting auxiliaries
selected from those now upon the market. In regard to effi-
ciency, most of the reflectors and globes which pretend to be
f
Room 3.
Room 2. Room 1.
Fig. 1. — Showing importance of shading lamp.
Room 3. Room 2. Room 1.
Fig. 2. — Showing variety of light distributions which may be obtained by use of ordinary reflectors.
Light. Medium. Dense.
Fig- 3- — Showing improved appearance when glassware is dense enough to conceal lamp.
Room 3. Room 2. Room 1.
Fig. 4.— Showing appearance with various wall decorations.
Room 3. Room 2. Room 1.
Fig. 5. — Showing appearance with various wall decorations.
Room 3.
Room 2.
Fig. 10. — Indirect lighting
MILLAR: ILLUMINATION OF INTERIORS IO7
efficient accomplish their purpose admirably. In fact so care-
fully has this element of the question been studied, that an ineffi-
cient reflector cannot to-day be successful unless it has some
compensating advantage, which renders it superior for some
purposes in spite of its inefficiency. I recently went through the
interesting experience of making a comparison of the standard
types of reflectors which were upon the market ten years ago,
and comparing them with types now available. In regard to
efficiency, the improvement has been very marked. Absorptions
of 10 to 20 per cent, now rule, where ten years ago absorptions
of 25 to 40 per cent, were typical for reflectors of substantially
similar light distribution characteristics. Improvement in the
quality of reflecting surfaces has gone hand in hand with im-
provement in the design of the curvature of such surfaces. As
in efficiency, so it is in appearance. Most of the reflectors of ten
years ago were opaque, and few were pleasing to the eye. To-day
even in the reflectors where efficiency in light re-direction is the
chief aim, pleasing appearance is the accepted order.
In the class of lighting auxiliaries in which decorative effect
is the chief object, a wide variety is available, and much of it
it pleasing and tasteful. Unfortunately, however, such auxili-
aries are characterized by inefficiency to an extent which appears
rather unnecessary. It is probable that in the developments of
the next few years, we shall note a strong tendency to improve
the efficiency of some types of decorative reflectors, without
interfering with their decorative qualities.
The influence of room decoration upon the amount of light re-
quired to illuminate a room properly is very marked ; or, stated
otherwise, with a given amount of light produced in a room, the
effectiveness of the illumination is largely influenced by the char-
acter of the decorations. Considering the simple case of a bare
lamp, employed to illuminate rooms having light, medium and
dark walls respectively, we may note a number of interesting
effects (Fig. 4). In the first place, the illuminated card on the
table in room No. 3 appears brighter than the cards in the other
rooms. It must be apparent that the card cannot be brighter be-
cause it receives light from the lamp and the ceiling only, while
the card in room No. 1, for example, receives light from the
corresponding light sources and ceiling which is enhanced con-
2
IOS TRANSACTIONS I. E. S. — PART II
siderably by light reflected from the walls. The card in room
No. i is actually 30 per cent, brighter than the card in room No.
3. That it does not so appear is an example of the effect of con-
trast, which in illumination is a very important fundamental. A
corresponding comparison may be made by observing the upper
part of the wall in each room, where again the white paper ap-
pears brighter in room No. 3 than in the other rooms. Though
actually not so bright as the white surfaces in rooms Nos. 1 and
2, these surfaces appear brighter in room No. 3 in comparison
with the dark wall-paper to which the eye naturally adapts itself
more or less.
In room No. 1, portions of the furniture which are but slightly
illuminated, as legs of the table, stand out distinctly, being sil-
houetted against the light rear wall. In room No. 3, so small is
the contrast between the rear wall and the dimly lighted portions
of the furniture, that it is difficult to discern the latter.
The glare due to the exposed light source is more serious in
room No. 3, due to the larger contrast between the light source
and the walls. Shadows of the furniture against the walls are
very prominent by contrast in room No. 1, in spite of the fact
that the shaded areas are more brightly illuminated by light which
is generally diffused within the room.
In considering this photograph, it should be remembered that
no light-directing auxiliary has been employed, and that therefore
a larger proportion of the light falls upon the walls than good
practise would dictate, if we except the darker room. In most
installations it is desirable primarily to secure the proper illumina-
tion of the lower part of the room, where the light is utilized, the
other requirements being that the ceiling and walls shall be
illuminated sufficiently to make the effect pleasing. When re-
flectors are used, the lighting effect of ceiling and wall decorations
is reduced greatly, if the reflectors are concentrating in character,
and reduced slightly, if they distribute the light rather broadly,
about the lower part of the room.
The brightness of walls is an important element, affecting
ocular comfort probably more seriously than the illumination of
the table plane. Generalizing, it is probably the best rule to avoid
extremes of wall decoration, whether they be light or dark. If
MILLAR: ILLUMINATION OF INTERIORS IO9
the walls are of high reflecting power, it is important to so direct
most of the lighting that the amount permitted to fall upon the
walls will not render them so excessively bright as to be trying
to the eyes. The illuminating engineer cannot control wall de-
corations, but he can control the light produced within the room,
and can so direct it as to secure the best effects.
In the next photograph a reflector which directs the light down-
ward rather largely is shown.
This detracts from the brightness of the upper portions of the
walls, the change of course being most apparent in room No. I,
where, due to the relatively high reflecting power of the wall-
paper the wall was brightest in the last photograph. The lower
portions of the walls are somewhat brighter than when bare
lamps were employed. Due to the better lighting of the floor,
the lower portions of the table and chairs, which with the bare
lamps could hardly be seen, are now slightly illuminated. With
this installation the effect of the ceiling and walls is lessened, be-
cause a smaller amount of light is permitted to fall upon them
reducing their illuminating power. That is to say, when a suit-
able reflector is employed, the table plane illumination intensity is
more nearly independent of reflection from ceiling and walls, and
instead of relying upon the latter for assistance in producing use-
ful illumination, the problem is simplified to one of rendering the
walls bright enough to produce a cheerful appearance.
It has been shown in the above that the ceiling and wall decora-
tions, when light in tone, may be of material assistance in increas-
ing the illumination intensity on the table plane. It may be argued
as the corollary of this, that when the walls are dark and incap-
able of augmenting the table plane illumination materially, the use
of reflectors for that purpose is all the more important.
I have discussed the effect of glare due to the presence of
a bright light source within the field of vision. This effect would
be almost, if not quite as disturbing, if instead of having a lamp
within view, an image of the lamp were to be seen in a mirror.
In that event, the effect would be due, not to the presence of the
light source, but to specular reflection of the light from the
mirrored surface. It is perhaps unfortunate that most artificial
surfaces which we are likely to view are sufficiently glossy or
HO TRANSACTIONS I. E. S. — PART II
polished to partake in some measure of the qualities of a mirror ;
that is, to reflect light specularly. Some surfaces which are very-
mat and free from gloss, diffuse the light so generally that the
specular element of the reflection is immaterial for most pur-
poses. But in most paper employed in books and magazines,
there is a considerable element of specular reflection, and this
characteristic is responsible for much of the difficulty which
demands adroit handling by the illuminating engineer in utilitarian
lighting.
Referring to the demonstration cards (Fig. 6), it will be
noted that the paper and letter on the right half of each have
glossy surfaces, while those on the left have diffusing surfaces
being almost totally free from specular reflection. On the right
half, one may see, when in line with the direction of the reflec-
tion, a distorted image, or a number of distorted images, of the
light source, much as though he were viewing the source through
a very imperfect mirror. On the left it is noted only that the
surface is illuminated and no trace of an image of the light
source may be seen. From all positions the letter on the left
half of the card may be seen. From a particular direction
(right photograph), that upon the right half of the card can be
seen only with great difficulty, if at all, because it is viewed from
the direction in which the glare is manifested.
No small part of the dissatisfaction with illumination installa-
tions is due to this effect of glare from observed surfaces. The
statement may be ventured also that no small part of trouble
with eyes is traceable to the same source. There are three
remedies : one is, to eliminate glossy surfaces wherever possible ;
particularly is this important in schoolbooks, and it is very grati-
fying to know that serious efforts are being put forth with a
view to regulating this matter. The second remedy is to reduce
the brightness of light sources as much as practicable by passing
the light through a diffusing medium of large area, or by reflect-
ing it from a diffusing surface of large area, in order that when
specular reflection from an observed surface is encountered, the
brightness of the light reflected may be so low as to minimize
the difficulty. The third remedy is to so locate light sources, or
to so locate the illuminated surfaces and adjust the position in
working or reading, that the direction in which light is reflected
MILLAR: ILLUMINATION OF INTERIORS III
specularly shall not be toward the eyes. All three of these
possible remedies should be kept in mind and applied wherever
practicable, and any one, or a combination of a part of each of
the three can be made effectual in reducing the trouble to a point
where it is not serious. The growing appreciation of the impor-
tance of this element of illuminating engineering work has been
the distinguishing feature of the past two years in the illumina-
tion field.
The oil lamp has in recent past years been the standard of
comparison for artificial illumination. Even to-day it is tradi-
tional among oculists that there is no artificial illuminant which
yields a light so free from detriment to the eyesight as does the
oil lamp. It is therefore of interest to note some of the condi-
tions under which the oil lamp has been used. Being essentially
a small illuminant, and both self-contained and portable, it was
natural that it should be placed close to the object viewed. This
entailed locating it more or less on a level with the eyes, and so
near to the observer that shielding the former became a matter
of natural course. In that fact is to be found the reason for
the development of the oil lamp shade. Given a small illuminant.
shaded for the protection of the eyes, there was no condition
under which visual difficulties could be experienced unless it
were attempted to read with the book in the illuminated zone
near the lamp base, with the reader facing the lamp (Room i).
Under such conditions, glare due to specular reflection from
the paper might be detrimental to vision, in which case it would
be so immediately apparent that instinctively the reader would
shift his position or the lamp slightly, in order to avoid it. With
light from this single light source incident upon the page from
a direction which would not result in serious glare from the
paper, and with the flame shielded from the observer's eyes, the
conditions for reading or other work were comparatively good.
At the same time, the old oil lamp was well adapted to the illumina-
tion of a book by light from over the reader's shoulder (Room 3),
one of the best positions for reading. It was in the comparative
freedom from misuse of the oil lamp, and the conditions which
its employment made natural, that the relative freedom from
harmful effect was probably found, if such freedom did exist
under those conditions, which is a point that has not been estab-
112 TRANSACTIONS I. E. S. — PART II
lished. Of course, no matter how favorable conditions for
vision may be, it is difficult to prevent a certain amount of care-
lessness or perversity in the use of the light (Room 2). Whether
it is carelessness which induces Mrs. Lux to read with her page
dimly illuminated by light from the ceiling and wall, while facing
the light, is open to question. Perhaps she may feel that her
appearance is more attractive with the light full upon her face.
Wall brackets may be employed with good effect if equipped
discreetly. For utilitarian purposes they are of value chiefly in
providing local illumination. It is very difficult to light a room
solely from wall brackets (Fig. 8). The light cannot be dis-
tributed satisfactorily in the room without placing light sources
immediately within the range of vision. Wall brackets find best
application when employed in rooms in which the main illumina-
tion is provided otherwise and the brackets are equipped with
decorative shades, the installation serving purposes of ornament
rather than utility.
Daylight, being that under which the human eye has been
evolved, may be expected to possess the qualities for which the
eye is best adapted. Neglecting other differences between the
natural conditions for which the eye is adapted and the artificial
conditions with which we have surrounded it, (such as the change
from distant to near vision and the change from use of the eye
during daylight hours only, to use of the eye for almost as long
a period during the hours of the night,) there still remain cer-
tain differences between artificial light and daylight, the study of
which forms a most interesting field for the illuminating engineer.
Daylight out-of-doors is the standard against which we must
compare both artificial light and daylight indoors, for the day-
light which is available in our interiors differs materially from
that out-of-doors in respect to quality, intensity and direction.
The intensities may be from 0.01 to 0.001 of those which prevail
out-of-doors in bright sunlight. The quality may differ not only
in respects which are not perceptible to the eye, but often differs
in color due to the influence of the absorption of colored walls,
etc., which materially alter the color of the natural light. The
direction is usually quite different. In regard to the desirability
of such direction of light as that which is prevalent in interiors
millar: illumination of interiors 113
illuminated by daylight, there is considerable discussion at the
present time, pro and con. My own view is that usually the
direction is undesirable. Coming through a window or windows
on one side, or at the most, two sides of a room, usually at an
angle somewhere between the horizontal and 45 deg. above the
horizontal, the light is very unequally distributed. The floor and
opposite wall receive the greater part of it, while the wall on the
side of the room in which the windows are cut, is illuminated only
by such light as may be reflected from the floor and the opposite
wall. Persons sitting in the room are likely to have the window
and the bright sky within the field of vision, or else they are likely
to sit in such a position that their faces are not well lighted. The
light is incident upon horizontal surfaces at a very sharp angle,
and there is only one good position for writing, or two good posi-
tions for reading, if glare from the window or the paper is to be
avoided, and as well shadow from the body or hand. Practically
the only way of bettering these conditions which has been
developed so far, is to utilize a window shade to protect the eye
against direct light from the sky, and this is done of course at
the expense of the illumination of the room. The usual direc-
tion of the light is in my opinion objectionable both from the
standpoint of utility and good appearance of the room. The
proper utilization of daylight for interior illumination is a sub-
ject of which the study has not yet been undertaken seriously.
There is one quality, however, in daylight, whether out-of-
doors or indoors, which has until recently been lacking in our
artificial lighting — and that is, ample diffusion. Interiors are
illuminated as a rule from a portion of the sky, the light source
being as large as the unobstructed portion of the window. Out-
of-doors, even in brilliant sunlight, the skylight is a considerable
factor in the total illumination. Of recent years more attention
has been given to this quality of diffusion, which previously had
been lacking in our artificial lighting. Early consideration of
lighting principles brought realization of the harm which exposed
light sources work, and led to attempts to conceal the light source.
There was evolved, among other systems, that of cove lighting.
In the process of concealing the lamps and permitting the light
to fall upon a white surface, from which a part of it was re-
flected into the room, the light was thoroughly diffused. This
114 TRANSACTIONS I. E. S. — PART II
system of lighting is more notable in regard to the success with
which it concealed the light sources and diffused the light, than
in other respects. Historically, it is notable for the evidence
which it affords of growing appreciation of some of the principles
of good lighting which are now considered to be thoroughly
established. The trouble with cove lighting as usually applied, is
that control of the direction of the light is lost, and that the flux
which is permitted to escape from the cove is diffused promiscu-
ously throughout the room, producing a flat and characterless
effect. Only a small portion of the flux is directed where it is
most wanted, while perhaps an equal portion is permitted to fall
upon surfaces where it is not desired in such quantities. The
system has not been largely applied, it being found possible to
realize its advantages by other methods which are free from some
of its disadvantages.
More recently another system of indirect lighting has been
developed, in which central fixtures are employed to conceal the
lamp from view and direct much of its light to the ceiling, from
which surface it is diffused downward. More engineering study
has been devoted to this system of lighting, and in consequence
its possibilities have been more largely realized, than were the
possibilities of cove lighting. This system of indirect lighting
has been widely exploited, and has given considerable satisfaction
in a wide variety of installations.
Direct lighting, in which the great bulk of the light utilized
comes directly from the light source, had been abused with
detrimental results. Particularly was it lacking in diffusion.
Indirect lighting is the other extreme, possessing in a high degree
the element of diffusion which is so often lacking in direct light-
ing systems. The rapid growth of indirect lighting is the mani-
festation of a protest against abuse of direct lighting. Its effect
has been to introduce into direct lighting practise a considerable
general improvement, which has corrected, or decreased some
of the evils of direct lighting. And too much credit cannot be
given to the exploiters of indirect lighting devices for the bene-
ficial influence which they have exerted upon our lighting prac-
tice in general.
In the lighting fixtures here shown (Fig. 10), the lamp in the
metal bowl is backed by an efficient mirrored reflector, which
m
Viewed from direction in which Viewed from direction in which
glare is not apparent. glare is apparent.
Fig. 6. — Showing glare due to specular reflection from glossy surface.
Good
reading posture
Undesirable
reading posture.
Fig. 7. — Showing manner in which a table lamp may be used.
Very
undesirable reading posture.
F'g- 8.— Illumination from wall brackets alone.
Fig. 9.— Daylight illumination.
Room 3.
Room 2.
Fig. 11. — Semi-indirect lighting
Room 1.
*•»
IBs v':
*^ -•- . .„ a - . 1
_ ; "j-Cy,
, ,.. -■■'■-—<
Indirect. Semi-indirect. Direct.
Fig. 12. — Three common methods of illumination.
Room 3. Room 2. Room 1.
Fig. 13. — Suggesting decorative or ornamental lighting units.
MILLAR: ILLUMINATION OF INTERIORS 11$
directs its light toward the ceiling. The rooms have been
equipped with three ceilings — one is white, and has about as
high a reflecting coefficient as is likely to be found in practise.
Another is cream colored, and reflects a smaller proportion of
the light. A third is dark cream, approaching a tan, and reflects
still less of the light. This latter is about as dark as one might
expect to find employed in an indirect lighting system, where any
attention is paid to efficiency. Indirect lighting is so largely
dependent upon the reflecting qualities of the ceiling, that the
statistics of the illumination intensities in these rooms are of
interest. The horizontal illumination intensity on the table plane
averages for the three ceilings :
White ceiling 100%
Light cream ceiling* 87 Jo
Dark cream ceiling 58 %
showing a reduced efficiency of 42 per cent, due to the inferior
reflecting qualities of the darker ceiling.
Following closely upon the development of the indirect light-
ing system, come systems classed inaccurately as semi-indirect
lighting units, in which part of the light is reflected from the
ceiling, as in the indirect system, while part of it comes directly
from the translucent bowl surrounding the light source. It is
obvious of course that with any translucent lighting auxiliary
employed in a direct lighting system, some of the light which
reaches the ceiling and walls, is reflected downward, and that
the system is thus a semi-indirect system. Those units which
are classed as semi-indirect units at the present time are, however,
units designed especially with a view to directing a considerable
proportion of the light toward the ceiling The most desirable
combination of direct and indirect light for general purposes
served by such units, is to-day a subject of discussion. Views of
illuminating engineers vary in this matter. All kinds of relations
between these two components are to be found represented by
outfits now available in the open market. These range from
equipments in which the transmitted light is so small a proportion
* It was discovered too late for correction that the light cream ceiling- has diffusing
qualities so unlike the dark cream ceiling that in spite of reasonably typical intensities
on the table plane, the appearance as viewed from without the rooms is not consistent
with the intensity figures shown ; thus, the ceiling in Room 2, when viewed from the
table, is much lighter than the ceiling in Room 3 though it does not so appear in the
figures.
Il6 TRANSACTIONS I. E. S. — PART II
of the total as to make it apparent that the purpose to be served
by the direct component is chiefly one of decoration, to those in
which the direct component is so large as to make evident an
intention to increase the efficiency considerably by restricting the
amount of light which is subjected to the inherent ceiling loss.
In the photograph (Fig. n) three semi-indirect lighting fix-
tures are shown. In room No. I, a direct lighting reflector is
inverted. In room No. 3, a bowl, not intended for this purpose,
is employed. The design of its surfaces is not well adapted to
this purpose, and it is therefore not so efficient as it might
otherwise be made. In room No. 2 a hemisphere is utilized,
illustrating semi-indirect lighting in the simplest of its charac-
teristic forms.
It is a matter for gratification that illuminating engineers
to-day have such an excellent choice as that afforded by the wide
range of available equipments for direct, indirect and semi-
indirect lighting systems. Each has its merits, each its demerits.
In some installations, one type is preferable, in other installations,
some other type may produce most desirable results. The good
qualities which characterize each are coming to be incorporated,
as far as practicable, in the others, and it may be noted that the
more vigorously each system is exploited, the more beneficial
upon lighting practise in general will the result be. With a
direct lighting system, it is a simple matter to direct a relatively
large percentage of the light downward upon say the table plane,
but it is a difficult matter to so dispose the lamps and to so
equip them that the installation will be free from troubles due
to glare and shadow. With an indirect lighting system it is
relatively a simple matter to avoid deleterious effects due to
glare and shadow, but it is very difficult to direct a satisfactorily
large percentage of the light upon the table plane. Where absence
of glare and shadow is a consideration of paramount importance,
an indirect or a semi-indirect lighting system may often be
preferable, in spite of the necessity for somewhat greater expen-
diture in energy. Where these considerations are not so im-
portant, or where economy of operation is the prime considera-
tion, a direct lighting system may prove preferable. In any case,
the adroitness of the illuminating engineer may exhibit itself in
miliar: illumination of interiors 117
securing the best balance between economy on the one hand
and absence of glare and shadow on the other. As to the appear-
ance of the installations, there may be all kinds of diverse views,
and we must remember that there is no disputing taste. Obvi-
ously it is difficult to discuss those phases of the question when
dealing with the subject in a general way.
The three modern systems of lighting are represented
in Fig. 12. In room No. 1, the direct lighting unit
transmits sufficient light to make the walls pleasantly, but not
objectionally, bright, while directing much of the light to the
table plane. In room No. 2, the semi-indirect unit illuminates
the card by light direct from the bowl and by light from the
ceiling and walls in something like the proportions of 3 to I. The
relative direct and indirect components upon the table plane are
of the order of iy2 to 1. In room No. 3 all of the light is
diffused from the ceiling. The ceiling is the brightest surface
within view, the lamp being entirely concealed. The illumination
is very soft and uniform.
Comparing the two end rooms, it will be noted that in room
No. 1, the vertical stripes in the wall-paper may be seen standing
out clear and sharp. The character of the pattern is evident. In
room No. 3 these stripes are seen somewhat less distinctly. This
is due to the lower intensity of light on the wall. Still more im-
portant, however, as a factor, is the downward direction of the
light from the ceiling. Viewed from the table, these stripes stand
out distinctly as the angle and direction are then such as to be
within the zone of strong specular reflection from the wall-paper.
Viewed as in the photograph, these stripes can hardly be dis-
cerned except on the upper part of the walls near the border.
The paper loses its character. This is an excellent illustration of
the importance of securing proper direction from the major part
of the light, although it should not be taken as an indication that
the direction is wrong in this installation because it must be
remembered that the effect would be minimized if the wall-paper
were viewed from within the room, instead of from without.
With the conditions as established (and it is not claimed that
they are more than suggestive of typical conditions) the card
illuminations are as follows :
Il8 TRANSACTIONS I. E. S. — PART II
Room No. I 220%
Room No. 2 100 Jo
Room No. 3 42 f0
It must be remembered however, that the direct lighting unit in
this case is favored, because the card is immediately beneath it
at the point of highest intensity. For purposes of reading, as an
example, it is difficult to judge from these figures as to the rela-
tive useful light. In the first place, questions of diffusion may
result in establishing demands for higher intensities in one system
than in another. This is one of the questions which is being
very generally investigated at the present time, and in such a re-
view as this, its discussion has no place. Dealing solely with the
question of distribution, it may be noted that most reading would
be likely to be done near the center of the room and that there-
fore the direct lighting system should receive some of the advan-
tage in rating which the high intensity of the card immediately
beneath the unit would appear to give it. The relative higher
intensities in the corners of the room with the indirect, and to a
lesser degree with the semi-indirect fixture, are not of much
advantage from a practical standpoint. In this particular in-
stallation, with the same flux produced by the lamps in each
type of lighting, the average horizontal intensities are relatively:
Direct lighting 1 .61 foot-candles
Semi-indirect lighting 1.33 foot-candles
Indirect lighting 0.91 foot-candles
It is generally believed that with conditions suitable for each
system of lighting, the direct lighting system will deliver about
twice as much light upon the table plane as does the indirect
lighting system, while each will illuminate the walls moderately.
The decorative feature has kept pace with developments in the
other branches of the art. Lighting auxiliaries are consistently
being improved in appearance, as well as in other features of
effectiveness. Efficiency of reflection, the necessary degree of
diffusion, and the proper direction of light are being achieved more
and more completely as experience becomes greater. In good
taste and other qualities that make for pleasing effects, constant
advances are being made also.
The older lines of lighting glassware, including the prismatic,
have been modified so as to render them more pleasing in appear-
millar: illumination of interiors 119
ance while the addition of a number of new lines of phosphate
and other glass affords the user a number of alternatives in the
way of glassware equipment suitable for use with any given type
of fixture in any ordinary installation.
In Fig. 13, may be seen illustrations of some of the more
decorative types of fixtures and glassware now available in stand-
ard types. Whatever the character of the installation may be, it
is more than likely that unless it is extraordinary, some fixture
and some kind of glassware may be obtained which may be used
in the installation with fair satisfaction. Unless installations are
considered which are so unusual as to demand the design of
special lighting equipments, those now obtainable must be con-
sidered to afford a very satisfactory range of selection.
The foregoing demonstrations must be taken with some quali-
fications. They do not pretend in all cases to be typical of any
particular class of installation. Time has not permitted a thor-
ough discussion of the characteristic qualities of any one of them.
Appearances have been different from those which would have
presented themselves had the rooms been observed from within.
The one thing which seems to me to have an immediate bearing
from the central station standpoint, and which I hope is re-
cognized is that if artificial lighting is to be made thor-
oughly good and satisfactory, it is necessary to thoroughly
diffuse and otherwise modify the light which is produced by the
lamps. This cannot be accomplished without considerable loss
of light, and therefore entails greater consumption of electrical
energy. Thus immediate commercial advantage goes hand in
hand with good business policy and with altruism, when the cen-
tral station spreads the gospel of good lighting among its
customers.
120 TRANSACTIONS I. E. S. — PART II
DISCUSSION AT A MEETING OF THE NEW ENGLAND
SECTION, MARCH 13, 1913.
Mr. A. E. Jossexyn : I think Mr. Millar's demonstration has
brought before us strongly the conclusion which probably many
of us had reached in investigating complaints of poor lighting,
viz., it is not so much a question of the light as it is of the
lighting, the kind of glassware, fittings or the radiation of the
light itself. It is not so much a question of the service supplied
by the lighting company. In my experience, covering several
years, we have many times been called upon to investigate what
bur customers claimed to be poor light and found, as a matter of
fact, that the cause was poor lighting rather than poor light;
that is to say the distribution of the light has been the cause of
the dissatisfaction. In many cases this has been due to the
location of the fixture or, if that has been located properly, it
has been due to the fact that the glassware was selected more for
decorative than for lighting purposes. I believe Mr. Millar's
paper has brought out the fact that it is not so much the light as
it is the lighting.
Mr. R. C. Ware : I think there is another point that is very
strongly brought out along the lines just mentioned. We get
complaints sometimes of poor light which is not only due to
poor distribution but to the fact that the customer has insisted
on using bare lamps. I think this very valuable demonstration
of Mr. Millar's has brought out very clearly the absolute necessity
of protecting the eyes ; and that bare lamps mean the stopping
down of the pupil of the eye to such an extent that the user
does not get the full benefit of the light actually given off. This
certainly ought to help us to talk intelligently and forcibly to
customers who complain that they do not get results. We ought
to be able to show them now why they do not, and to help them
on to the road so that they may get what they are after.
Mr. J. W. CowlEs: I judge that others are affected in the
same way that I am to-night, in being somewhat over-awed by
this paper in its many opportunities for discussion. What we
have seen and heard has been brought out in such rapid succes-
sion that for my part I am quite bewildered in knowing just
where to enter into the subject from a discussion standpoint.
ILLUMINATION OF INTERIORS 121
There is so much that we have seen, — so many points have been
brought out in a vivid and interesting manner, — that a con-
siderable amount of careful thought is required for the proper
digestion of the material offered. I have not as yet had an
opportunity to read this paper in print, but my firm resolve
to-night is to sit down with this paper in the quiet of my room,
go through it carefully, and gain from it the profit that must
come from the perusal and more gradual study of it.
Prof. George A. HoadeEy : I think one thing in the indirect
lighting system should be taken into account, that is, the eye
accommodates itself to the light in the room; you get the im-
pression when first coming into the room, that the illumination
is insufficient, but after having been in the room a short time,
the illumination becomes sufficient. I might cite an example of
the lighting of a dining-room in which an inverted cone shade
was used which gave a spotty light on the table, and after
putting a piece of ground glass across the bottom of the angle
shade, the illumination became satisfactory.
122 TRANSACTIONS I. E. S. PART II
DISCUSSION AT JOINT MEETING OF I. E. S. NEW
YORK SECTION AND NEW YORK COMPANIES'
SECTION OF THE NATIONAL ELEC-
TRIC LIGHT ASSOCIATION,
NOVEMBER 18, 1912.
Dr. Herbert E. Ives: The demonstration which Mr. Millar
has given us this evening is, in my opinion, one of the most
instructive ever given on the subject of illumination. We all
realize from his use of dolls and toy furniture that this talk was
meant to be of a kindergarten nature. Nevertheless I feel sure
that I am speaking for all present, even those who in the words
of the chairman have spent a lifetime studying illumination, when
I say that we have learned a very great deal from his clear
presentation and admirable demonstration. I wish to make no
criticism of this, for it deserves nothing but praise.
Of course no one studying such a comparatively new subject
as lighting will agree with everyone else. I would like to take
this opportunity to emphasize a point which Mr. Millar could
not, in the time at his disposal, treat. I want to call attention to
the fact that in all these demonstration booths the light source is
a centrally overhead fixture. We are so accustomed to such a
system of lighting that I think we are apt to overlook the fact
that it is not the only possible method and perhaps not even the
best. The other evening there was a paper presented at the
Philadelphia Section of the Illuminating Engineering Society on
indirect lighting, and one remark made by the speaker was to
me very suggestive. He said that all of the fixtures used were
placed on the outlets which had been "planned by the contractor."
It is a question in my mind whether we have not reached the
point where we must go beyond the contractor and his ideas.
For instance, may it not be possible that the light of the future
will be from the side rather than from overhead? The lighting
of a room by daylight is from the side and is generally considered
pretty satisfactory. It differs, too, from most artificial lighting
in the size of the light sources.
Mr. Millar has followed the usual classification of direct, indi-
rect and semi-indirect lighting. To my mind, however, the
proper classification is on the basis of the size of the sources.
ILLUMINATION OF INTERIORS 123
We have been accustomed to small light sources which are
necessarily of high intrinsic brilliancy, and we are now working
towards larger sources of lower intrinsic brilliancy, whether it
be by the use of translucent diffusing media or by diffusive
reflecting material on the ceilings. In the illumination of most
rooms by daylight we have a very large light source, namely, the
sky, which at the same time is usually not visible to a person in
the room who looks out at the neighboring houses or at the
lower portions of the landscape. The net result is illumination
from a large invisible light source from which the general direction
of illumination is at the side. I hope to see experiments made
with a view to meeting these conditions by artificial light. I
feel sure they will be instructive and they may lead to some
satisfactory systems of artificial lighting.
Right in line with Mr. Millar's concluding remarks I may say
that according to all present indications if we do copy daylight
illumination in the manner I have suggested it will mean the use
of an enormously greater amount of electrical energy or gas.
Dr. C. H. Sharp: I feel very much as Dr. Ives that there is
much to be said in the way of praise about what we have seen
here to-night. I have known for some time that Mr. Millar had
in preparation a demonstration of light in miniature, and I have
been wondering, without knowing anything about it, what he
could make out of it. How could he show to us lighting effects in
rooms which we are not inside of but which we are merely allowed
to look into and give us any adequate idea of what is really the
effect inside those rooms ? My questions have all been answered.
I can say that he has been able to show us a great deal regarding
practical conditions of illumination.
I agree with Dr. Ives regarding daylight illumination.
I think it is pretty good, and one reason why it is
good is because there is plenty of it, and if there is
not plenty of it it is not good. Under the usual New York con-
ditions we are not often much affected with the glare from the sky.
We do not see the sky, but we do get a thoroughly diffused light
in the room which is sufficient in amount but which is directed.
Xow I think that it might be possible to work out something in
artificial illumination to simulate daylight illumination. Imagine
3
124 TRANSACTIONS I. E. S. — PART II
you have around each window of the room, say, a trough re-
flector with a lot of lamps so directed that if they were turned on
they would throw light out of the room. Now when night comes
on and light ceases to stream in through the windows, we draw
down over the window a very white window shade, opaque, and
with a very good white surface, and then turn on the lights. Then
the lamps will throw a strong illumination on the window shade
and which will throw it back into the room directed and diffused
and distributed similarly to daylight. There should be less glare
than there is in daylight. This would have another advantage in
that if the office furniture and the arrangements of the room in
general were made so as to be most advantageous for daylight
illumination, they would also be the most advantageous arrange-
ments for the night time. I have not made any computations on
this as to the number of watts per cubic foot it would require, but
I have no doubt that the plan would be practicable in certain cases
even if the efficiency were not very high. It could hardly be more
extravagant than other indirect systems.
Mr. L. B. Marks : As I heard Mr. Millar talking this even-
ing I recall a letter which I received about two years ago from
Dr. Ives in which he suggested that I make similar demonstra-
tions at the Johns Hopkins University in Baltimore in connection
with my lectures on principles and designs of interior illumina-
tion; but I was afraid to try it. I did not have the nerve Mr.
Millar has. As I sat here this evening and heard his lecture and
viewed the demonstrations of the principles discussed by me at
Baltimore, I felt convinced that anybody who had the good
fortune to see the demonstration which Mr. Millar has prepared
would receive an object lesson in illumination which I believe
would be far better than any theoretical discussion of the subject.
With regard to the question that was brought up this evening
of producing a distribution and direction of light similar to that
of daylight, I have a somewhat different view than that expressed
by the other speakers. It seems to me that it is desirable to have
a change at night. We all want a little variety. We do not want
to eat the same dish at every meal, breakfast, lunch and dinner.
For daylight illumination we must depend almost exclusively
on side windows. It is a fact that we cannot plan the best day-
ILLUMINATION OF INTERIORS 125
light illumination because of the physical limitations of buildings.
We usually have a number of floors in each building and cannot
get a desirable distribution of light in all of them, if indeed in
any of them. We have not that limitation at night. Then why
not avail ourselves of the broad scope of application we have in
electric and gas lighting? If you discuss this matter with the
architect or with the decorator he is likely to tell you he can
obtain much more pleasing effects at night by artificial light than
he can in the day time, because he has the ability to place the
light sources where he wants them. One of the criticisms which
Mr. Edward Caldwell the distinguished fixture designer and
decorator made of my lectures was that I did not lay sufficient
stress on the importance of changing the character of the
illumination at night. In his opinion that constituted one of the
important things that an illuminating engineer has to study up —
the arrangements of various lights and shades and the production
of pleasing effects at night, not possible by daylight.
There is another matter which I want to speak of briefly,
namely, Mr. Millar's closing remarks with reference to the
greater use of light. In the I. E. S. Primer which was distributed
this evening the principles underlying nearly all of Mr. Millar's
demonstrations are discussed and illustrated. The keynote of
this Primer is good diffusion of light. What does good diffusion
of light mean ? It usually means more light and better light. I
hold that it is up to the illuminating engineer if he finds it neces-
sary to recommend more light to do so and to do so fearlessly,
even though the customer is told he must pay more money for it.
One of the first things that the lighting company's representative
will be up against when the public calls for greater diffusion of
light is the complaint that the lighting bills are larger. We have
seen here to-night a demonstration which will convince any man —
and what is more important any woman of the household — that
it pays well to use more current or gas for lighting if you can
get better results in diffusion of light.
It would not be a bad idea to get out a miniature exhibition set
of this kind that the salesman or company's representative could
take with him to make a demonstration to the housewife.
Mr. D. McFarlan Moore: Mr. Millar has laid before us
a wonderful wealth of ideas. We can hope for some future
126 TRANSACTIONS I. E. S. — PART II
paper to carry out in greater detail some of the modifications of
his general scheme that suggest themselves. But I admire the
thoroughness with which he has worked out the equipment details
of his three demonstration rooms.
In these days we hear a great deal about direct, semi-direct and
indirect lighting. Still better diffusion is continually desired.
Indirect lighting virtually consists in changing a point source to
the entire ceiling as the source of light.
I have been introduced as the inventor of the Moore light. We
should not forget that one of the solutions of this problem of
better diffusion is to directly increase the area of the light source
itself, abandoning the intense light of bulbs for the soft light of
long tubes, thereby also avoiding the necessity for any form of
reflecting or diffusing or softening glassware, but using the light
directly at the intensity at which it is generated.
Dr. A. S. McAllister : I have nothing to say in addition to
the remarks, except that I wish to compliment Mr. Millar on his
excellent demonstration, the thoroughness of it, the accuracy and
fineness of the work, etc. The subject of light from the aspect
which Dr. Ives spoke of is one to which I have given some little
thought. I do not quite agree with him that we want the light on
the side. In order to get it on the side we must place it in the
line of view. Now I do not care how much it diffuses, we must
have a little of that, and that to my mind is not objectionable. I
am inclined to think that light above is oftentimes more
advantageous.
Mr. C. A. Ljttlefteld : I wish to express to Mr. Millar
personally and on behalf of the section my appreciation
of the wonderful demonstration he has given us this
evening. Its chief advantage is that it is so complete
and comprehensive, yet so simple and devoid of unin-
teresting technical detail. To my mind one of the most success-
ful things the Illuminating Engineering Society has done, from
a popular standpoint, is the publication of the Primer. If in
some manner this lecture and the Primer could be made integral
parts and be presented in various parts of the city and country
a vast amount of good could be accomplished. We all know
that very little is known of correct lighting principles, and even
ILLUMINATION OF INTERIORS 12"J
in my own home I know that lighting conditions are not as they
should be, but could this lecture be given before popular audi-
ences, in churches, lecture halls, etc., it would get before the
public generally a knowledge of correct lighting fundamentals
that would do an enormous amount of good. It is too good to
be kept merely for technical societies and meetings, and I trust
that some way may be found whereby this plan can be carried
out. I feel that this must have been one of the motives that
animated Mr. Millar to give such an enormous amount of his
valuable time to prepare this lecture. Mr. Millar is too well
and favorably known among the profession to need any eulogy
from me, but a public delivery of this lecture would much
enhance his reputation and at the same time do a great amount
of good.
Mr. Norman Macbeth : There is very little I can add. I con-
sider the demonstration Mr. Millar has made here to-night one of
the finest things I have seen in a long time. His methods of show-
ing the wonderful detail work of the various sections of these
rooms, the smoothness of the stage work, and his lecture itself
are remarkable. I believe a great many people here have an
entirely different idea of the lighting question after seeing Mr.
Millar's demonstrations.
Mr. A. J. Marshall : I wish to express my congratulations
to Mr. Millar on his most clever, instructive, and entertaining
demonstration, which I feel privileged to have been able to
witness.
I am a very great believer in catering to the brain through
the eye in educational work, and consequently am the better
able to appreciate the character of Mr. Millar's elaborate and
painstaking experiments.
It would seem unpardonable to criticise the effects that we
have witnessed this evening on account of their general all
around value. However, I would like to state that none of the
interiors which we have seen, to me represent good practise in
the lighting of living rooms. As a matter of fact, I certainly
would not employ any that I have seen. Most of the effects
shown are those associated with the past or with what might
be termed inappropriate "mechanical" lighting equipment, which
128 TRANSACTIONS I. E. S. — PART II
certainly does not represent good practise, or which has the sup-
port of architects, fixture houses, up-to-date central stations, gas
companies, the public, etc. In fact, it is just the equipment as
shown that advanced workers are eliminating.
This, however,, should not be considered as being uncom-
plimentary to Mr. Millar in any way, shape, or form. The criti-
cism is directed simply to equipment which is to-day not con-
sidered good practise.
The interesting phase of the whole matter is that something
has been done, and Mr. Millar should be complimented upon his
success in doing that something exceedingly well.
Mr. A. W. Stark: I am very glad to tell you that I was
very much impressed with Mr. Millar's lecture. I think, how-
ever, that Mr. Millar omitted a very important subject in con-
nection with his lecture, and that was the quality of the light.
I agree with Mr. Marks, who spoke a moment ago, that we need
a variety at night. If it is possible to distribute light in a better
manner than comes through window openings from different
floors of the buildings that we have to occupy through the day,
such an arrangement is much more desirable. However, the
quality of the light is vitally important. In looking around I
think it safe to say that 25 per cent, of those here are wearing
glasses. Why? I think that if we had somebody present who
knows more of the subject than I do he would probably say they
have misused their eyes. I think that fact is generally due to
not only misusing the eyes, but that the quality of light was not
best for the eyes. Daylight will not affect as seriously, if
properly used, the eyes, as any artificial light. I would have
been very much more pleased with what Mr. Millar had to say
if he had given us a demonstration of what gas light would do.
Mr. C. B. Graves : The question of indirect lighting and low
ceilings probably enters more into residential work than in any
other work. It is simply a matter of distribution, and in that
case the fixtures must be suspended within a distance of some
14 or 18 inches from the ceiling. Very good results are obtained
from low ceilings, as it simply means a matter of a larger
number of fixtures for the area to be covered. I think that where
there is room for clearance, that is, where a ceiling of &l/2 to
9 feet high is available, that indirect lighting can be worked in
ILLUMINATION OF INTERIORS I2Q,
very successfully, and in fact it has been done in a great many
cases.
Mr. Preston S. Millar (In reply) : I could myself also
mention a few things that have been overlooked in these demon-
strations. We have the necessary equipment. We have a de-
finite schedule for about 40 demonstrations, and we have pre-
sented 12 in a period of one hour and a quarter; 40 would take
six hours. We did not present the others.
130 TRANSACTIONS I. E. S. — PART II
ILLUMINATION AND EYESTRAIN.*
ELLICE M. ALGER, M. D.
Synopsis: In this paper the following topics are discussed in the
order given: Nature of eyestrain and the part played by illumination —
Daylight the ideal illuminant — The composition of artificial lights and
the effect of the short waves in producing discomfort and disease in the
eyes — The intensity of light and its lack of diffusion as factors in ocular
fatigue and inefficiency — The relation between poor lighting and industrial
accidents — The arrangement and position of lights and their influence on
efficiency — Experts agreed on the general principles involved in good
lighting, but not on the details — Education and discussion must precede
any drastic legal regulation.
Eyestrain is the common expression for that rather compre-
hensive group of symptoms which result from abnormal ocular
fatigue. It results from compelling eyes to do work which is
beyond their physiological capacity. Things close at hand are
seen by a muscular effort of focussing which, when long con-
tinued, produces a normal fatigue and requires a definite period
of recuperation. If the eyes tire sooner than they should because
of some intrinsic weakness of ciliary muscles, or because of a
handicap imposed by astigmatism or increasing years, the fatigue
is apt to manifest itself not only by defective vision but by pain,
and we speak of the condition as an accomodative asthenopia.
The eye likewise sees things through the effect of light falling on
a sensitive retina. If this light be over bright, or if the retina
by reason of over-exposure or disease is hypersensitive, the
result is the disturbance of vision and pain which we call retinal
asthenopia. The results of eyestrain are manifold and affect no
two people exactly alike. They include pains in the eyes and many
functionl defects of vision and, quite possibly, often result in
organic eye disease as well. They cause 80 per cent, of the
chronic headaches. They often result in functional disturbances
of other organs and in conditions of general nervous exhaustion
and irritability. While most of the symptoms that come from
eyestrain are of the accomodative sort, these are all capable of
aggravation by improper lighting and there are so many that are
* A paper read before a meeting of the New York section of the Illuminating Engi-
neering Society, March 13, 1913.
AivGER: illumination and eyestrain 131
caused directly in this way that I shall invite your attention for
a time to the relation between eyestrain and illumination.
In studying this relation it must be remembered that there are
few exact standards of ocular capacity. The average individual
can see objects of a definite size at a definite distance and this
average is taken as a standard, but there are many who fall below
this standard without obvious cause, and many who are far above
the average. The variations in muscular endurance are still
wider and one man can work hour after hour at tasks which
fatigue another in a very short time. The sensitiveness to light
likewise varies widely in different individuals, both ability to see
distinctly by faulty light and ability to work without exhaustion
in strong light.
Light is the reaction excited in the retina by the impact of
certain vibrations or waves in the ether, which cause different
sensations according as they are longer or shorter. The long
ones give the sensation of red light, while, as they get shorter
and shorter, one may see in succession all the colors of the visible
spectrum. The mixture of all these wave-lengths together pro-
duces the sensation of white light. But the visible spectrum does
not include all the waves by any means. There are longer waves
than the red which cannot be seen but can be felt as heat, and
shorter ones than the violet which have a very active chemical
effect. It must be remembered too that both these qualities
exist in the visible spectrum, the heating effects predominating
at the red end, while the violet end approaches the ultra-violet in
its chemical activity.
This enables one to explain some of the untoward effects of
daylight on the eyes, even though daylight affords the best illumi-
nation for ordinary purposes.
Many of the effects of sunlight which were once attributed to
heat are now known to be due to chemical activity. For instance,
in snow and desert blindness the light is broken up by reflection
from the crystalline snow or sand, and the actinic waves produce
intense inflammation of the conjunctiva which, if long continued,
results in total disability. Even in temperate climes one suffers
more or less from glare and burn from direct or reflected sun-
light, and by common consent a good north light is taken as the
standard of ideal illumination, being the steadiest, the pleasantest
I32 TRANSACTIONS I. E. S. — PART II
to the eyes, the best diffused, causing the fewest shadows and
affecting color values least.
An artificial light can be broken up into its component parts
and its spectrum compared with that of daylight, and its illuminat-
ing power can be measured by aid of various photometers ; but
so far there is no artificial light which is just like daylight, though
we are said to be getting nearer and nearer to it.
It has been shown by experiment that the light which gives the
maximum of illumination with the minimum of irritation of the
eye is composed of the yellowish rays from the middle of the
spectrum. For this reason the old fashioned candle and kerosene
lights have never gone entirely out of fashion. But most of the
more recent artificial lights, whether gas or electric, contain a much
higher proportion of the short violet or actinic rays and some of
them contain many of the ultra-violet rays as well. When un-
shaded their chemical activity is so great that they can be used
for various therapeutic purposes. They are capable of tanning
the skin, and of causing symptoms like those of a modified snow
blindness. Prolonged exposure to the electric arc light some-
times produces an intense conjunctivitis with contraction of the
pupils and erosions of the cornea, which fortunately generally
yield readily to treatment. Nearly everybody has experienced the
discomfort and premature fatigue that comes from reading by un-
shaded incandescent lights. Even if they do not actually produce
inflammatory changes themselves, they certainly render those al-
ready present decidedly less tolerable.
It is quite possible, however, that the delayed actinic effects of
light whether natural or artificial are much more serious. The
ultra-violet rays are arrested by ordinary glass, and in the eye by
the tissues of the cornea and lens so that the deeper structures of
the eye escape harm, but there is strong reason to suspect that
their constant absorption by the lens may be one of the causes of
cataract. Experimenters have been able to demonstrate lenticu-
lar changes in the eyes of rabbits exposed to such lights; and it is
known that stokers and glass-blowers, who have to face very
brilliant incandescent light,, have a tremendous predisposition to
cataract. This so called bottle-makers' cataract begins not in the
anterior part of the lens, which would be expected if heat were
ALGER : ILLUMINATION AND EYESTRAIN 133
the essential factor, but in the posterior portion where the rays of
light are most concentrated. Other suggested observations are
that in the ordinary cataract of old people the first changes gen-
erally occur in the lower inner quadrant of each lens, which is the
part least shaded by the brows and so most exposed to sunlight
from above, and that when cataracts develop in people who have
one light and one dark eye, it invariably appears first in the one
unprotected by pigment from the light.
Even if the ultra-violet rays do not reach the deeper structures
of the eye, one must not forget that the shorter waves of the visible
spectrum have decided actinic properties. Many people have had
their eyes permanently ruined by incautious watching of an
eclipse, and similar damage sometimes follows exposure to electric
flashes and even to long exposure to the arc light. In such cases
the light is condensed on the surface of the retina resulting in
local inflammation and degeneration, that particular spot be-
coming permanently blind.
Oculists suspect, though they cannot prove, that less intense and
longer continued light irritation may be a factor in many similar
degenerative changes in the retina and chorioid, and advise both
for prophylaxis and treatment the use of amber glasses, and
shades of such composition as to soften the light and exclude the
actinic end of the spectrum. To people who are at all sensitive
to light they are a great comfort.
Our north light is soft and even and well diffused, so that it
causes a minimum of shadows. Artificial light to give anything
like the same amount of illumination must be much more con-
centrated and intense. Now the human eye even in natural light
has to adapt itself to so many variations of intensity and dimness
that it has developed a very beautiful mechanism for regulating
the amount of light admitted to the retina. When the light is
dim the pupil dilates, and when it is bright it contracts sharply. A
sudden very bright light causes pain not because the retina hurts
but because of this sudden extreme muscular contraction of the
iris. Constant exposure to bright light necessitates constant muscu-
lar contraction and engenders in many people premature fatigue.
Still more tiresome and painful is the rapid dilation and contrac-
tion of the pupil that results from the varying intensity of a
134 TRANSACTIONS I. E. S. PART II
flickering light. Furthermore, intense and long continued ex-
posure to bright light causes retinal exhaustion and the retina is
capable of reacting only to powerful stimulation. In other words,
that retina becomes for the time being blind except in the brightest
of lights. Every one has experienced the comparative blindness
caused by going from bright sunlight into a dimly lighted room.
It is a common experience to have workmen insist on having as
intense a light as possible because they have temporarily so
blunted their retinal sensitiveness that they are helpless without
it, and it is generally the hardest kind of a task to convince them
that even if they suffer no harm from the glare they cannot
possibly work as long without fatigue. In another set of people
the retina instead of being blunted becomes hyperesthetic and
finally almost incapable of bearing any exposure to light at all.
This condition is seen at its worst in hysterics, when it is of course
not a result of over lighting, but there are a number of occupa-
tions like those of the gilders and polishers who have their atten-
tion fixed for long periods on bright surfaces, in which retinal
asthenopia is very common.
Furthermore, daylight has a vast volume and is dif-
fused so that objects get light from all sides, and shadows
are reduced to a minimum. Artificial light can hardly
be expected to secure thorough diffusion and more than compara-
tive freedom from shadows, but in many industries almost no
attention has been paid to this point. And yet it is very im-
portant, for Calder in a very interesting paper has shown that
the retinal anesthesia and deep shadows that result from poor
artificial lighting are potent factors in causing industrial accidents.
The records of some 8,000 manufacturing plants over a period
of three years showed a regular minimum of accidents during
July and August which gradually increased to a maximum in
the dark winter months. The influence of daylight in preventing
accident was much more evident in occupations which require
not so much bright light as diffused light without shadows over
large areas as in the building trades for instance. Indoor work-
ers as often suffer accident from too much light as from too
little. Exact photometric measurements often show that the light
of ordinary incandescent lamps concentrated at the cutting- point
ALGER : ILLUMINATION AND EYESTRAIN 135,
of a tool or a work-bench is often several times the intensity of
daylight. But the eye adopts itself to this intensity and when
the workman turns from his over-lighted work, perhaps in a
room full of moving machinery, he is practically blind. What is
needed from the point not only of safety but of health and com-
fort is much less intensity and much better diffusion of light.
This applies to all walks of life. We have all become accustomed
to using far more intense light than we need.
One can measure the amount of illumination, by photometers
which are much more accurate and dependable than the human
eye; but, after all is said and done, the eye is one of the best of
photometers if one is careful not to injure it in the process, since
it is upon its adaptability to that eye that all artificial lights must
stand or fall. One should begin with a low illumination and
gradually increase it till a point is reached when further increase
ceases to improve the details of the work in hand. Beyond this
all additional light is both unnecessary and physiological extrava-
gance.
Abnormal fatigue is admittedly one of the greatest predis-
posing causes to most diseases be they physical or mental, and
though the part played by bad lighting is perhaps not clear cut
it is beyond doubt. In most factories, schools and offices the eyes
must be used constantly for work of a character they were never
intended for. The result even in normal eyes is a muscular and
nervous fatigue which is measurably increased by both over or
under-lighting. The first engenders fatigue from retinal exhaus-
tion and pupillary spasm, while the second results in the strain
that follows sharp focussing and constant attention. In the
majority of individuals whose eyes are handicapped by astig-
matism or other refractive errors, the strain is still greater. I
shall not take your time with the long list of conditions of health
which have been attributed to eyestrain. Some of these are
beyond dispute, others are still in question.
The over-lighting which is so common to-day may conceivably
have other effects. Woodruff has shown that in the tropics
blondes who are unprotected by skin pigment are over-stimulated
by the bright light and finally develop a characteristic nervous
exhaustion. It is quite possible that eyestrain and the constant
exposure to intense light of short wave-length may be predis-
136 TRANSACTIONS I. E. S. — PART II
posing factors to the neurasthenia from which our garment
makers admittedly suffer.
The arrangement of lights as well as their composition and
intensity is of importance. It is well known how uncomfortable
it is, and how much it interferes with clear vision to have a bright
light shining directly into the eyes, and lights which enter the
eye from below are much more annoying than those from above.
And yet how often are machines so placed that the operator has
to face a window or a light. The same difficulty occurs in trades
like those of the gilders and polishers who have bright lights
reflected into the eyes from their work, and in schools where
the smooth shining pages of the books answer the same purpose.
So far as possible light should fall from above, behind, and to
one side. The light should be sufficient for the work in hand,
should throw no shadows on the work, and should be reflected,
not into the eyes of the workers but to one side. When it
comes to the arrangement of light for many workers in a factory
or school the problem is very much more difficult and presents
many technical details, which must be left in the hands of the
illuminating engineer.
Even when estimated by its actual cost in dollars and cents, bad
lighting is often more expensive than good, but from the stand-
point of efficiency there is no comparison. Bad lighting un-
doubtedly causes unnecessary strain of the eyes and consequent
premature muscular fatigue; it compels closer and more con-
stant attention to the details of work, so that tasks which
should be done almost automatically and without mental
effort are done consciously. Under such circumstances
the output of each individual is manifestly less than it should be,
there is a larger percentage of mistakes and material spoiled, and
the number of accidents, large and small, is vastly increased.
Even under the best of conditions, the extreme subdivision of
factory work with its consequent monotony, largely destroys the
pleasure of work, but bad eyes and poor lighting and long hours
are important factors in the industrial discontent of the day.
It goes without saying that any system of scientific shop man-
agement worthy of the name implies a good lighting system as
one of the first requisites, but as yet opinions vary widely as to
just what this means. It is possible to regulate the color and
ALGER : ILLUMINATION AND EYESTRAIN 137
composition of the light that enters the eyes by the interposition
of screens or shades which shall absorb the rays one does not
wish to use, or by having it reflected from suitable colored sur-
faces. The volume and intensity of light can be regulated by
increasing or diminishing the number of units, and by diffusing
it with frosted shades, or by reflecting it from rough surfaces.
But while the experts are agreed on the principles involved they
do not agree entirely on the details. The human eye is flexible
enough to adapt itself to very wide variations in illumination but
there must be comparatively narrow limits within which the
greatest efficiency may be reached. Quite possibly different
industries may require entirely different types of illumination
and while these may be worked out in detail in the laboratory
they must all be subjected to the final test in the shop or school.
Illumination as a science is yet in its infancy. Even in great
public buildings, libraries, and theatres it is treated not as an
essential but simply as an aid to the proper display of the genius
of the architect or the taste of the decorator. And if such build-
ings are badly done one can hardly expect as yet that any great
attention will be paid to the proper lighting of the ordinary
factory or house.
Every one admits to-day that the state must control factory
conditions so far as they effect the health and well being of em-
ployees and many attempts are being made both here and abroad
to deal with the subject of illumination by law.
It is an extremely difficult subject to handle in this way; even
the experts are not agreed on many important points. What
would be good lighting in one industry might be the worst possi-
ble in another. To make drastic regulations in the present state
of the art would often involve manufacturers in great expense in
changing their light equipment without any guarantee that it
would be permanently satisfactory.
After all, good lighting is essential to the efficiency of both
employer and employees, and a judicious campaign of education
will make them both appreciate it. Then too, there are numerous
large and powerful corporations engaged in various branches of
the lighting industry and there is perhaps little danger that the
subject will be allowed to be forgotten either by consumers or
legislators.
I38 TRANSACTIONS I. E. S. — PART II
DISCUSSION.
Dr. Percy W. Cobb (communicated) : Dr. Alger has said that
there are few means of measuring ocular capacity, to which I
would add "which are of practical use in the question of illumina-
tion."
To the ophthalmologist the test-types are familiar, by which
the size of the smallest type that can be read furnishes the basis
on which visual acuity is measured. Among other means of
estimating the capacity of the eye, one is by its perception of
small differences in brightness in fairly large surfaces — to be dis-
tinguished from visual acuity estimation, where the brightness-
difference between object and background is large and the size
of the object seen is minimal. Similarly the color sensitivity of
the eye can be measured and many other criteria of the sensitive-
ness of vision have been worked upon, but never made practical
for the purposes which concern illuminating engineers. That
branch of physiology is as yet in its infancy.
The major disturbances of the eyes, such as ophthalmia elec-
trica, snow blindness, occupation cataract and the like, probably
due to the ultra-violet radiation, are however quite frank and to
the ophthalmologist easily recognizable. A noteworthy point is
that these disturbances all have their origin in conditions far re-
moved from those of customary illumination. On the other hand,
the troublesome minor and more familiar disturbances which
appear in the use of ordinary illuminants (headache, smarting of
the eyeballs, blurring of vision and so on) seem to take place
under conditions such that physical considerations practically pre-
clude the ultra-violet radiation as the cause ; for we know that the
light from an ordinary tungsten filament is, for equal visible light,
far poorer in the ultra violet light than daylight, and is used at
illuminations almost incredibly lower than those that obtain in
daylight conditions. With a good daylight illumination in a
room, turning on the artificial lights makes almost no impression
on the eye, indicating how relatively low the artificial illumina-
tion really is.
It is another matter when we come to consider such a thing as
a light source in the visual field. Here we have an intense focuss-
ing of energy upon a minute area of the retina, and a disturbance
ILLUMINATION AND EYESTRAIN 139
soon results, as anyone knows who has glanced at a naked fila-
ment for a second or two and noticed the after images of it which
subsequently disturb his vision. Whether the ultra-violet plays
a part, this is an open question. We must remember that the
preponderance of the visible radiaton is extremely large in cases
which the illuminating engineer has to consider.
Bearing on this point is some work quite recently published by
Dr. Ferree, from which it appears that the eye suffers a consider-
able loss in power from work under a system of artificial
illumination, of which a number of the sources come within the
field of vision. The loss in the power of the eye appeared not in
the acuity itself, as the eye was able to see equally small test-
letters after the period of work, but in its ability to maintain that
power; the identical test-letters viewed after the work-period
showed a greater tendency to appear blurred at times.
Apart from high intrinsic brightness of objects it seems much
more than probable that widely unequal illuminations in different
parts of the room can induce retinal disturbance and eye-strain.
The writer is inclined to explain the minor eye-disturbances just
mentioned largely on this basis. The eyes working in an over-
done local illumination, on a bright page or a machine with its
numerous reflections, are turned to relatively dark places, for
which they are wholly unadapted, and back again to the intense
light — for which by that time they are again unadapted. It is not
hard to imagine how such conditions can induce retinal fatigue
and muscular eye-strain. A more tangible loss due to such a
system is the economic value of time wasted in fumbling for a
needed article in a dim light with bright-adapted eyes; and above
all the danger of accident, especially in manufacturing plants, to
persons obliged to move about in relatively dim surroundings in
which their eyes, recently under high illumination are more or
less blind.
For such reasons legislation as to illumination conditions in
factories, schools and other public and semi-public places is an
important part of the modern movement for the conservation of
human resources, and it can be expected that such legislation will
progress whenever experts in illumination can make it clear ex-
4
140 TRANSACTIONS I. E. S. — PART II
actly what is the most efficient lighting, when all things are con-
sidered.
Dr. Sinclair Tousky (communicated) : The paper is re-
markably complete and accurate.
Eye-strain is much more commonly a result of imperfection
in the eye, than in the light. The muscles which regulate the
size of the pupil for the amount of light, act automatically
and practicably without fatigue and provide for a wide
latitude in the intensity of illumination. Where eye-strain
is diagnosed it is not enough to correct errors in illumination, but
the eye itself should be tested as to accommodation or focussing
power, as to astigmatism or difference in curvature and refractive
power along different meridians of the eye, and as to the ocular
muscles which should be so accurately balanced as to automati-
cally direct both eyes toward the same object without fatigue.
The ultra-violet rays are much more abundant in sunlight upon
the mountain top, than in our cities where they have been filtered
out by passage through additional thousands of feet of air and
especially the dust in the air. Filtered water is more transparent
to the ultra-violet ray than the air we breathe.
Dr. Alger is right in recommending glasses as a protection
from sun-light upon snow-fields and in the desert and for electric
arc lights ; and it should be added that the worst accidents come
from the blinding flash when an electric current is short-circuited
and that these workers also should be protected by glasses.
Glass is practically opaque to the ultra-violet rays, but quartz
or "pebble" lenses are transparent to them. Colorless glass is a
complete protection from the invisible ultra-violet rays but of
course if the luminous rays are too powerful, the colored glasses
which give the greatest sense of rest, amber color perhaps, may
be used.
I use glass in the measurement of ultra-violet content in light
used for the therapeutic effect of those rays. The light from a
mercury-vapor arc in a quartz tube may produce a certain effect
upon a photographic film in a small fraction of the time required
to produce the same effect through a thin piece of glass. Such
a light will quickly produce a severe burn upon the healthy skin ;
and cures the terrible disease called lupus, but is manifestly un-
ILLUMINATION AND EYESTRAIN I4I
suited for ordinary illumination. A mercury- vapor light in a
glass tube generates an equal quantity of ultra-violet rays but
they cannot escape from the tube and have no harmful effect.
The light from any electric arc is rich in ultra-violet rays and
should always be filtered through a glass shade for illuminating
purposes.
In applying my test for ultra-violet rays, it must be understood
that no camera lens is used, the film or sensitized paper is ex-
posed directly to the light. And any light which produces a
markedly greater photographic effect through a sheet of quartz
crystal than through a sheet of glass is regarded as too rich in
ultra-violet rays to be desirable for illuminating purposes.
Professor Alger quotes Woodruff's statement that it is not the
heat but the light in the tropics which injuriously affects the white
races. This theory has been very widely tested and the general
conclusion is against it. So that instead of the dark underwear
advised by Woodruff, white clothing has been found to be better.
One other point should be mentioned and that is the eye-strain
which inevitably results from a flickering light.
Eye-strain may frequently be prevented by the presence of a
dimly lighted back-ground upon which the vision may rest dur-
ing moments when the person looks away from his work. Con-
stant staring at work at a fixed distance and with either too
much or too little illumination, will cause strain in the best of
eyes.
Dr. C. E. FerrEE (communicated) : Dr. Alger's paper is in
brief compass an excellent and interesting symposium of the sub-
ject. The present writer has, however, the following comments
to make.
(1) On the third page of his paper Dr. Alger says: "It has
been shown by experiment that the light which gives the maximum
of illumination with the minimum of irritation to the eye is com-
posed of the yellowish rays of the middle of the spectrum. For
this reason the old-fashioned candle and kerosene lamp have
never gone entirely out of fashion." Although he willingly leaves
himself open to correction, the present writer does not believe
that the above conclusion can be justly drawn from the experi-
mental evidence in existence at the present time. There are two
142 TRANSACTIONS I. E. S. — PART II
points to Dr. Alger's comparison : "maximum illumination" and
"minimum irritation." Of these two points, so far as the writer
knows, definite experimental work has been done only on the
former.1 By "maximum illumination" the writer asumes from
the relation of the statement to the general subject of lighting
that Dr. Alger means the maximum of illumination for seeing
detail or maximum acuity of vision.2
There are two ways in which acuity of vision may be con-
sidered in relation to the problem of lighting: acuity as deter-
mined by the momentary judgment, and acuity which represents
the average of ability to see detail for a period of time. Acuity
as determined by the momentary judgment does not show the
progressive loss of efficiency resulting from a period of work
even under an unfavorable lighting system, because under the
spur of the will the muscles of the eye, though they may have
lost enormously in efficiency, may be whipped up to their normal
power long enough to make the judgment required by this visual
acuity test. Acuity, then, as determined by the momentary judg-
ment can only be used to determine the general level or scale of
efficiency for the fresh eye. It can not be used to detect loss of
efficiency. In the problem of lighting, however, the general level
or scale of efficiency as determined by the momentary judgment
of acuity is, comparatively speaking, of minor importance. What
is needed is a type of illumination that gives the highest average
of acuity or efficiency in seeing for a period of work and at the
same time the least loss of efficiency. So far as its relation to
the quality of light is concerned, tests have been made up to the
present time only of acuity as determined by the momentary judg-
ment. With regard to this type of acuity Dr. Alger's statement
1 Although no definite experimental work has yet been done on the effect of varying
the quality of light on its tendency to produce discomfort, still it can be said from the
results of our own work that when intensity and distribution are equalized, an installa-
tion of clear carbon lamps, which gives a light comparatively rich in yellow and red, shows
a greater tendency to produce discomfort than an installation of clear tungsten lamps,
the light from which contains a proportionately greater number of the short wave-
lengths.
- Visual acuity as usually tested involves the discrimination both of visual angle and
of brightness difference. As it enters into seeing in ordinary life, it may involve also the
discrimination of differences in color quality.
ILLUMINATION AND EYESTRAIN 143
will be here examined in the light of the work done by three men
selected as typical : Langley,3 Luckiesh,4 and Rice.5
Langley made his determination of acuity with the colors of
the spectrum equalized in energy. Luckiesh and Rice, on the
other hand, worked with colors equalized photometrically. Some
again have exercised no especial intensity control at all. In the
writer's opinion, Langley's conception is the correct one. We
want to know for equal outputs of energy what color gives the
greatest acuity of vision. To equate the colors photometrically
is to equate them for seeing, which in a measure begs the ques-
tion at the outset.0 Langley, working with the light of the
spectrum, showed that for equal amounts of energy (radiometri-
cally determined) the maximal acuity of vision is given when
light in the region of the green and blue-green is employed. His
results, therefore, give no support to the belief that yellow light
possesses an advantage over lights of other colors for clear seeing.
Luckiesh was not concerned primarily with making a com-
parison of the different colors for acuity, although such a com-
parison may be made from his results. His problem was to show
that colors taken from a narrow region of the spectrum give
greater acuity than colors more complex as to wave-length, and
to find an explanation for this phenomenon. An examination,
3 Langley: Energy and Vision, American Journal of Science, 18S8, XXXVI, 3rd Ser.,
PP- 359-379-
4 Luckiesh, M.: The Influence of Spectral Character of Light on the Effectiveness of
Illumination, Trans, of the Illuminating Engineering Society, 1912, VII, p. 135-158.
6 Rice, D. E.: Visual Acuity with Lights of Different Colors and Intensities, Arch, of
Psychol., 1912, No. 23, p. 1-59.
See also Uhtoff: Archiv fur Ophthalmologic 1S86, XXXII, (1), p. 171. A. Konig: Zeitsch.
f. Psychol, u. Physiol, d. Sinnesorgane, 1893, IV, p. 241; and Sitzungsbcr d. Berliner Akad. d.
Wissensch., 1897, XIII, p. 559; Pfliiger: Ann. d. Physik, 1902, IX, p. 185; Oerum: Skandi-
navisches Archiv fur Physiol., 1904, XVI; Boltunow: Zeitsch. f. Psychol, u. Physiol, d.
Sinnesorgane, 1907-8, XLH, (2), p. 359; Broca and Laporte: Bulletin de la Societe Inter-
nationale des Electnciens, Paris, 1908, VIII; 2nd Ser., No. LXXVII; Dow, J. S.: London III.
Eng.. II, p. 233; Ashe, S. W. : Electrical World, Feb. 25, 1909.
6 That is, visual acuity as ordinarily tested involves a discrimination of brightness
difference as well as of visual angle. The discrimination of blightness difference sustains
a relation both to the degree of illumination and to the color quality of the light. Up to
a certain point an increase in the degree of intensity of illumination increases the dis-
crimination of brightness difference. The presence of dominant color in the light, on the
other hand, interferes with this discrimination or masks the difference. If, then, the
colored lights are made of the same degree of luminosity as determined by the photo-
metric judgment, there remains only one of the above factors to make them differ in the
degree of acuity they produce, namely, the effect of color quality on the discrimination of
brightness difference. Therefore, I have said that to make a comparison of visual acuity
for the different colors with lights equated photometrically is in a measure to beg the
question at the outset.
144 TRANSACTIONS I. E. S. — PART II
of his curves show a greater acuity for yellow than for any of
the other colors of the spectrum. But a comparison can not be
drawn from his results of acuity for yellow and white light; it
can be made merely for yellow and the other colors of the spec-
trum. Moreover it is far from safe to pass from the results of
his experiments to the conclusion that yellow light gives greater
acuity than white light; and still less safe from a literal inter-
pretation of these results to conclude that white light with yellow
as its dominant hue, such as is given by the kerosene flame, gives
greater acuity than is given by clear white light. Yet we presume
that either results of this kind or results of photometric observa-
tions which strictly speaking are not applicable to the point in
question, are responsible for the belief, somewhat generally held,
that yellow light possesses an advantage for seeing over white light.
Theoretically speaking, this belief might at first thought seem
to have considerable justification, for if white is made up of all
the colors, and of these colors yellow gives the greatest acuity, then
yellow should also, it might seem, give greater acuity than white
light of equal luminosity. Too much would, however, be taken for
granted in drawing such a conclusion, for as stated in the pre-
ceding foot note, visual acuity, as ordinarily tested, involves a dis-
crimination of brightness difference as well as a discrimination
of visual angle, and the presence of a dominant color
in the light strongly interferes with this discrimination. In
Luckiesh's experiments this color factor was present in case of
each of the comparisons made. Moreover, it was probably
weakest in case of yellow, for yellow is the least saturated of the
colors of the spectrum. Hence, yellow in this regard possesses
an advantage over all the other colors for the clear seeing of de-
tails executed in white and black, as is required in the visual
acuity test. In the case of yellow vs. white light, however, the
advantage is reversed. No color factor is present to reduce
acuity in the determinations for white light.7 In short, there-
7 The difference between-the results of Langley and L,uckiesh was doubtless due to the
difference in their method of equating their colors. Since Langley equalized his colors in
energy, he had differences both in luminosity, and in saturation and quality of color to
affect his discrimination of the brightness difference between his test object and its back-
ground for the different colors; while Luekiesh, since his colors were equated in lumin-
osity, had only difference in color quality and saturation to affect this discrimination.
Possibly resolving power of the lens should also be included as one of the factors influ-
encing acuity for the different colors; but since the resolving power according to Ravleigh
is greatest in the blue, resolving power could not at least have been a dominating "factor
in I,uckiesh's observations.
ILLUMINATION AND EYESTRAIN I45
fore, to reason without further experiment from results of the
kind obtained by Luckiesh, to what would happen in case a com-
parison should be made for any of the colors and white light, as
apparently has often been done, is to pre-suppose the assumption
of a degree of simplicity with regard to the eye factors that does
not exist.
Rice did not work with the light of the spectrum. He used the
rougher method of isolating his wave-lengths afforded by color
filters. He determined the acuity for light approximating white,
(given in some cases by a carbon, and in some cases by a Nernst
filament), and for the colors given by his red, green, and blue
filters. These colors were made photometrically equal, each to
each and to the white light. Ten different degrees of intensity
were employed. The acuities for these lights were found to be
in order from greatest to least : white, red. green, and blue.
Yellow did not enter into the comparison.8
Even, then, with regard to acuity of vision as determined by
the momentary judgment, it has not been established by experi-
ment, so far as the writer knows, that either monochromatic yel-
low or white light predominantly yellow possesses an advantage
for clear seeing over clear white light ; and with regard to a
visual acuity determination which represents an average for a
period of work, no experiments have as yet been made which
can be regarded as determining the effect of quality of light on
clear seeing. In his own experiments the writer has planned to
determine in order the effect of differences in distribution,
intensity, and quality of light, both on the power of the eye to
hold its efficiency during a period of work, and to maintain its
maximum state of comfort. The effect of differences in dis-
tribution and intensity is now being worked out. The results for
distribution have in part been published,9 and more will soon be
8 Oerum (op. cit.) also made a comparison of acuitv for white light and the colors.
He found white light to give the greatest acuity. Of the colors, red, green, and blue, red
gave the greatest: green next; and blue the least. Boltunow (op. cit.), however, making
a similar comparison, found green to give the greatest 'acuity of these three colors, and
red the least. White gave a greater acuity than an3' of the colors. Ashe (op. cit.) de-
termined the acuity for red, green, blue, and white lights of equal luminosities. He
found the greatest acuity for white lights. Of the colors, blue gave the greatest acuity,
green next, and red the least. Dow (op. cit.) found that light in the region of the blue-
green gives the greatest acuity for near objects, and light in the region of the red for dis-
tant objects.
9 C. E. Ferree. Tests for the Efficiency of the Eve, etc. Trans. I. E. S. (Jan., 1913),
vol. VIII., p. 40.
I46 TRANSACTIONS I. E. S. — PART II
published. A systematic study of the effect of differences in
quality has, however, not yet been undertaken. The writer is,
therefore, not at this time in a position definitely to commit him-
self on this point. He can say, however, from the results of the
work already done, that with distribution and intensity equated,
an installation of clear carbon lamps, which gives a light rela-
tively rich in red and yellow, causes the eye to fall off more in
efficiency as the result of a period of work than an installation of
clear tungsten lamps, the light from which is whiter and contains
proportionately more of the shorter wave-lengths. In short, it is
the writer's contention that the question whether or not white or
colored light is better for the eye can not be answered until definite
tests are made of this point alone under conditions in which all
other factors are rendered constant. The merits of the kerosene
flame, for example, as compared with other sources of illumina-
tion, must be tested under a system of installation that gives the
same intensity at the source and, as nearly as possible, the same
distribution in the field of vision as is given by other illuminants.
This has not been done at all. Our judgment of the comparative
merits of the color quality of the light given by it are based on
the roughest kinds of impression obtained under conditions of
installation in which there has been no attempt at control of the
other factors that influence the effect of light on the eye.
(2) On the sixth page of his paper Dr. Alger says: "One
should begin with a low illumination and gradually increase it
until a point is reached where further increase ceases to improve
the details of the work in hand." That is, visual acuity is here
made the test of the amount of light that should be employed.
But, as is stated above, visual acuity, that is visual acuity as
determined by the momentary judgment, is not the only or even
the most important factor that has to be taken into account in
lighting. The element of time must be introduced into the test.
That is, an intensity of illumination must be chosen at which the
eye holds its maximum acuity or efficiency for a period of work.
This is by no means in every case the same degree of illumination
that gives the maximum of acuity as determined by the
momentary judgment. For example, our tests for loss of
efficiencv for different intensities of illumination with a given
ILLUMINATION AND EYESTRAIN 147
type of installation do not show that the degree of illumination
that gives maximum acuity for the momentary judgment gives
also for every kind of installation the least loss of acuity or the
maximum average of efficiency for a period of 3-4 hours of work.
Moreover, the comfort of the eye must also be taken into account.
Here also results are wanting to show that the degree of illumi-
nation that gives maximum acuity as determined by the
momentary judgment gives also maximum comfort. In short,
the degree of correlation between visual acuity, loss of efficiency
and the tendency to produce discomfort can not be taken for
granted. These factors constitute three separably determinable
moments, no one of which should be neglected in installing a
lighting system.
(3) Dr. Alger says on the eight page: "The human eye is
flexible enough to adapt itself to very wide variations of illumina-
tions, but there must be comparatively narrow limits within which
the greatest efficiency may be reached." Our work on the effect
of varying the intensity of illumination on the eye's loss of
efficiency shows that in general this statement is true. The range
of favorable intensity varies widely, however, with the type of
installation. It is, for example, much narrower for the direct
than for the indirect lighting system.
(4) On the seventh page Dr. Alger says that "lights which
enter the eye from below are much more annoying than those
from above." This statement is also borne out by the results of
our experiments. In making a preliminary study of the causes
of discomfort, a light of constant intensity was thrown on the
retina at different points in its several meridians, and the time
limen of discomfort was determined. This limen was found to
be lower for the upper than for the lower half of the retina; and
for the nasal than for the temporal half.
(5) On the fourth page Dr. Alger says: "A sudden very
bright light causes pain not because the retina hurts, but because
of a sudden extreme muscular contraction of the iris. Constant
exposure to bright light necessitates constant muscular contrac-
tion and engenders in many people premature fatigue."
Magendie's experiments in 1824 showed that the retina and optic
nerve are insensitive to pain from mechanical stimulation. These
I48 TRANSACTIONS I. E. S. — PART II
and similar experiments have led to the belief that the discomfort
experienced on exposing the eye to a degree of illumination to
which it is not accustomed is muscular. That it can not be
wholly or even essentially muscular is shown by the fact that it
is gotten in cases where the ciliary and iris muscles have been
paralysed by atropin; also in cases where the lens has so long
been removed that muscular atrophy must have taken place. In
short, there is no doubt in the present writer's mind that the dis-
comfort experienced as the result of work under unfavorable
conditions of lighting is not by any means all muscular. The
"sandiness" passing over into a stinging stabbing pain which
comes early in the experience of discomfort seems to be con-
junctival. Just what other reactions come as the result of ex-
posing the retina to a degree or kind of illumination to which it is
not accustomed is for future work to determine. That they
can not all be muscular is plainly obvious.
hunter: home lighting 149
HOME LIGHTING.*
BY GEORGE LELAND HUNTER.
Synopsis: The following paper presents some of the elementary con-
siderations of residence or home lighting, particularly from the viewpoint of
the interior decorator. The author regards home lighting as a decorative
problem. He discusses in a general way the influence of colored surfaces
and contrast upon the appearance and the decorative scheme of interiors. He
suggests the co-operation of the architect, illuminating engineer and
decorator to insure success and progress in lighting problems of this kind.
In the houses and apartments of New York and other large
cities there are thousands of rooms gloomy and cheerless because
they are not properly lighted, either by day or by night. Because
of such rooms hundreds of flats and apartments are unrented
which if properly illuminated could be quickly leased to respon-
sible and permanent tenants.
The first impulse is to rush to the lighting fixture man or to
the illuminating engineer, for assistance in solving the problem.
From the name, people naturally assume that an illuminating
engineer is one capable of producing ideal conditions of illumi-
nation in any kind of an interior. As a matter of fact the illumi-
nating engineer has not always a remedy at hand to correct the
bad lighting of dark rooms and apartments. Nor is the lighting
fixture man any better off. They can talk about fixtures, candle-
power, frosted bulbs and diffusing shades and the distribution
of outlets, etc., but when assigned a problem in lighting a
dark room agreeably and effectively they often fail. This is not
their fault. The specialization of industry prevents most light-
ing fixture men from knowing anything about interior decoration
and furnishings.
The problem of illuminating the interiors of houses and apart-
ments is not only a fixture, but a decorative problem. Success
in the future lies in the co-operation of architect, decorator and
lighting fixture man. Of course when the architect really under-
stands the art of interior decoration, as well as the planning and
construction of buildings, his is the master mind that can best
guide the lighting man and the decorator.
* A paper read before a meeting of the New York section of the Illuminating Engi-
neering Society, December 12, 1912.
150 transactions i. e. s. — part ii
Illuminating a room means making clearly visible the form
and color and texture of the objects in it. If the walls, furniture
and other objects in a room are properly illuminated, then the
room is illuminated. If they are not properly illuminated, then
the room is not properly illuminated.
Of course there is opportunity for all kinds of contrast in
lighting a room. Often the woodwork — the architectural back-
ground, the frame-work of the room — can be accentuated by
being kept a little heavier in color and consequently darker than
the flat surfaces of the walls. Often too the furniture is very
properly accentuated in the same way. But if all the objects
in a room, as well as the walls and ceiling and floor, are dark
and somber, it is then impossible to illuminate the room at all.
The only thing that can be done is to distribute around the
room a large number of light units having comparatively large
shades of diffusing glass that will give large surfaces of bright-
ness— but surfaces that are not too bright. With this wide dis-
tribution of light sources, wherever the eye turns it meets bright-
ness, so that the effect is one of brightness. In illumination that
is everything.
In measuring illumination, the photometer is practically use-
less. The human eye is the only capable judge. Whether a
room is photometrically bright is of not the slightest importance.
The only thing of importance is "Does the room look bright?"
If it looks bright it is bright.
It is a well known fact that light surfaces reflect much light
and that dark surfaces reflect little light. This means that the
dark surfaces absorb the light, while the light surfaces reflect it.
This applies, of course, not only to the walls of a room but also
to the windows of a room. In illumination this fact is of vital
importance. If the walls of a room are light, they not only
look bright but they also reflect light to other walls, which
re-reflect the light. And at each reflection but a small part,
comparatively, of the light is lost.
I have noticed that many people in the evening leave the
window shades up. Apparently it does not occur to them that
every open window is a transparent hole through which the light
leaks. And light costs money. Dark window shades also absorb
hunter: home lighting 151
a great deal of light. Light shades not only reflect light in the
room but by their brightness help give an appearance of bright-
ness and cheerfulness to the whole interior.
While it is true that light surfaces always tend to recede from
the eye and dark surfaces to advance toward the eye, it is also
true that small spots or objects tend to stand out against a con-
trasting background. But bright objects against a dark back-
ground tend to stand out less than dark objects against a light
background. These facts are of prime importance in the decora-
tive and economical lighting of a room. The only way to light
residences economically is to light them decoratively. If they
are not lighted decoratively they are not lighted at all. It is
lighting that makes a room agreeable and comfortable to live in.
The shape and size of rooms are also of vital importance in
illuminating them. I find that a great many lighting fixture men
do not seem to have any definite relation between the shape and
size of rooms and the amount of light they provide. I will admit
that the amount of light necessary varies very greatly according
to the way the rooms are finished and furnished. A room fin-
ished and furnished in dark colors with surfaces of rough texture
might require ten times as much light as a room with light smooth
surfaces, and not be illuminated at that.
In my own practise, in rooms 9 feet high, I allow a 16 candle-
power lamp for each fifty feet of floor space. In higher rooms
I allow 10 per cent, additional light for each increase of one
foot in height. This will be found too much light for light
rooms, but too little for dark room.
There are some rooms in a house with ceilings too high for
their size — such as narrow halls, bath rooms, etc. In these rooms
ceiling reflection should not be utilized. The moment a ceiling
is brightly lighted it appears to rise from one to two feet. Con-
sequently in bath rooms and narrow halls the light should be
kept off the ceiling, and placed upon the side walls, making
the side walls appear to recede while the dark ceiling appears
lower.
Of course, if a room is very wide and low, it is desirable to
use ceiling reflection. By making the ceiling bright, its height
is apparently increased ; at the same time it may be used as a
secondary distributor of the light.
152 TRANSACTIONS I. E. S. — PART II
Most of the talk about indirect illumination is nonsense. I
would absolutely bar from use for ceiling fixtures anything that
does not allow a fair proportion of the light to come down
through the fixture. The best fixtures for what is called semi-
indirect lighting are alabaster bowls and bowls of a similar kind
of glass.
And now for the solution of the problem of lighting dark
rooms in flats and apartments — lighting them by day as well as
by night — even when the only window is on an air-shaft. The
air-shaft should be light in color, white or ivory; then its walls
will pick up and transmit the maximum of light from above and
outside, provided there is a lateral opening. The only thing to
do in lighting a room thus opening on an air-shaft is to make
its walls and ceiling and furniture light in tone. Then in the
daytime they will pick up from the air-shaft all the light there
is and by reflection and re-reflection will give an appearance of
cheerfulness and comfort to the interior.
One is often amazed to learn that while some owners of apart-
ments have been fully instructed on this point they fail to act.
The reason is as follows, as I learned by questioning several
janitors. To finish the wood-work and doors of a small room
in dark tones with cheap varnish costs about 75 cents and the
varnish lasts on an average about three years ; so that the real
expense per room is only 25 cents. To finish the same room in
white or ivory enamel costs $2.00 and the ivory enamel often
needs renewing every year. Consequently the apartment owner
balances $2.00 in one hand against 25 cents in the other, and says
he will stick to the cheap varnish.
In decorative and economical lighting it is best to favor gold
and yellows and oranges and greens of light tone, at the expense
of reds and blues, particularly these in dark tones. Dark red
and dark green or blue shades on lamps or fixtures simply kill
the light. Especially effective in toning light for residence use
are ground-glass shades of various shapes and sizes. These not
only eliminate the burning effect of the light source, but they
also cream the white slightly by eliminating some of the blue.
I dare say that by this time I have made it clear why I regard
the problem of illuminating residences and apartments as a prob-
home: lighting 153
lem for the decorator and architect rather than for the illuminating
engineer. But I want to say right here that unless the decorator
and the architect take lessons from the illuminating engineer
they will make terrible mistakes.
DISCUSSION.
Mr. George S. Barrows: Mr. Hunter's paper presents a part
of the subject which many, who are qualified to solve the prob-
lems ordinarily presented, may find unfamiliar; and, therefore,
for the best interest of the ultimate consumer, they should prob-
ably consult with some one familiar with that side of the question,
which they have not studied in detail — the architect or decorator.
It is quite likely that neither the architect, the decorator nor
the illuminating engineer would be quite satisfied with the solu-
tion of the problem solely by the other one, and therefore a
compromise must be effected so that the result will be neither
decorative alone on the one hand nor utilitarian alone on the
other hand.
In one town where I am somewhat familiar with the method
of operation of the illumination experts of a large gas company,
the salesman endeavors to meet the architect and have him
design the fixtures, suggesting, however, the number and the
location of the outlets. The architect also indicates the design of
the glassware and the general color of the light which he thinks
should be used to best emphasize the details of the decoration. It
is then up to the illumination expert to see that the proper
burners, mantles and glassware are used, and if, in his opinion,
the requirements of the architect will not give satisfactory results,
a conference is arranged in order to discuss and eliminate the
points of difference.
Mr. G. B. Nichols : This subject is one of very vital interest
to the Illuminating Engineering Society at the present time, for
to obtain the highest efficiency in any system of illumination as
planned for a new undertaking, there must be this co-operation
of all concerned in designing the entire structure.
In general the owner holds off in making up his mind to build
until the last moment or until the money needed has been entirely
raised, and then he wishes building construction to start at once.
This gives the architect but very little time to prepare his plans
154 TRANSACTIONS I. E. S. — PART II
so as to meet the owner's wishes or so that building con-
struction can be started at the proper season of the year. After
the architect has the plans well under way (in most cases
about eighty to ninety per cent, completed) he, or the owner
himself, will then engage specialists to co-operate in planning
the different branches of engineering work required, such
as illuminating, electrical, heating, ventilating, sanitary, etc.
The architect is so rushed at this last stage of the work, that
minor details such as color and character of finish of walls and
illuminating fixtures are wholly deferred until after the building
is well under way and more time can be given to their considera-
tion. The illuminating engineer therefore must in most cases
plan the outlets without knowing very much about the details
of the rooms aside from general shape and use. Furthermore,
the character of the illuminating fixtures is seldom considered at
this time, the illuminating engineer being informed to provide
enough outlets and sufficient wiring to take care of any system
afterwards required.
A great deal of the trouble now causing poor illumination
can be traced directly to the fixture houses who have not in the
least tried to adapt their material and machinery to the present
standards now being adopted, but rather force a great many of
their antiquated and inefficient fixture designs to the front without
any regard for illumination, or to the expense of upkeep of them.
In summing up it can truly be said that the architect needs
the illuminating engineer to guide him in the selection of the
appropriate illuminant; the illuminating engineer needs the archi-
tect's advice to design the proper standard or hanging for the^
illuminant and to approve the location and type of same so as to
harmonize with the architectural design of the building, and at
the same time the architect and the illuminating engineer need
the skill of the decorator to blend the colors and wall decora-
tions so as to add to the dignity and efficiency of the architect's
design and the illuminating engineer's conception of light and
shadows. The fixture manufacturer should also lend his aid in
giving ideas in adapting the fixture design selected to the most
economical method of manufacture so as to get the results both
in respect to design and illumination at the lowest cost.
TRANSACTIONS
OF THE
Illuminating Engineering Society
Published monthly, except during July, August, and September, by the
ILLUMINATING ENGINEERING SOCIETY
General Offices: 29 West Thirty-Ninth Street. New York
Vol. VIII
APRIL, 1913
No. 4
Erratum.
The illustration at the bottom of
page opposite page 106 of March, 1913
(Vol. VIII, No. 3) Transactions has
the words "light" and "dense" trans-
posed. The globe on the left should
have been designated as dense, and the
one on the right light.
Council Notes.
The April meeting of the council was
held on the nth instant, in the general
offices of the society, 29 West 39th
Street, New York. Those in attendance
were: Preston S. Millar, president;
L. B. Marks, treasurer; C. J. Russell,
George S. Barrows, W. J. Serrill, J. D.
Israel, general secretary; V. R. Lau-
singh, Norman Macbeth, Alan Bright
(representative of Mr. Howard S.
F,vans, vice-president of the Pittsburgh
section) and James T. Maxwell and
H. E. Ives by invitation.
A monthly report on the finances and
membership of the society was received
from the assistant secretary. Accord-
ing to the report of the current assets
of the society, as of April 1, was
$6,818; $4,313 of that amount repre-
sented cash in bank; the remaining
$2,505 represented accounts owing to
the society. The unpaid bills, April 1,
amounted to $775. The net current
assets as of that date, therefore,
amounted to $6,043. The expenses for
the first three months of 1913 was said
to have aggregated $1,821. The mem-
bership, including the additions and the
defections presented at the meeting,
totaled 1,338 members. The membership
at the beginning of the year was 1,335
members.
Vouchers Nos. 1241 to 1272, inclusive,
aggregating $1,033.70, which had been
approved and submitted by the finance
committee, were authorized paid.
Mr. L. B. Marks, chairman of the
committee on illumination primer, re-
ported that the primer was to be trans-
lated into French by Professor A.
Blondel of Paris and published within
the near future. It was understood that
the translation by Professor Blondel
would not be published by any com-
mercial organization. It was resolved
that a set of the electros of the illus-
trations in the primer be sent gratis to
Professor Blondel with the suggestion,
that, if at any time they should be
utilized for publication of the primer
by any commercial organization, the
society would desire to be reimbursed
for their, cost.
The executive committee was em-
powered to undertake preliminary nego-
tiations looking toward a course of
lectures on architecture and decoration
with special reference to illumination.
A monthly report on activities of the
TRANSACTIONS I. E. S. — PART I
Philadelphia section was received from
Mr. W. J. Serrill, vice-president.
A brief report on the work of the
Chicago section was received from
Vice-president Cravath.
President Millar announced a number
of additional appointments to several
committees. These were approved.
The officers of the society were em-
powered to appoint from time to time
representatives in cities not having sec-
tions of the society.
The executive committee was em-
powered to act and fix a time and place
for the 1913 convention of the society
upon recommendation of the committee
on time and place for the annual con-
vention.
A list of members who were delin-
quent in the payment of their 1913 dues
was read.
Section Notes.
CHICAGO SECTION
The April meeting of the Chicago
section was postponed. Notices of the
May meeting will be published shortly.
NEW ENGLAND SECTION
The New England section held a
meeting in the auditorium of the Edison
Building, Boston, April 21. Dr. Louis
Bell presented a paper entitled "Notes
on the Effect of Radiation on the Eye."
NEW YORK SECTION
At a meeting of the New York sec-
tion in the United Engineering Socie-
ties' Building, April 9, Mr. M. Luckiesh
of the National Electric Lamp Associa-
tion, Cleveland, presented a paper en-
titled "Light and Art." The paper was
supplemented by a series of demonstra-
tions showing the effect of the direction,
the quality and the distribution of light
on various art objects. Mr. J. B. Taylor
of the General Electric Company, Sche-
nectady, N. Y., presented a paper on
"Color Photography" which was also
supplemented by a series of lantern
slides and demonstrations. About 135
members and guests were present.
PHILADELPHIA SECTION
The Philadelphia section held a joint
meeting with the Philadelphia Gas
Works section of the National Com-
mercial Gas Association and the Phila-
delphia Electric Company section of the
National Electric Light Association,
April 23, in the Parkway Building,
Broad and Cherry Streets, April 23.
Mr. T. W. Rolph of the Holophane
Works of the General Electric Com-
pany read a paper on "Metal Reflectors
for Industrial Lighting."
PITTSBURGH SECTION
At a meeting of the Pittsburgh sec-
tion, April 18, in the Oliver Building,
Mr. J. Frank Martin of the Duquesne
Lighting Company presented a paper
entitled "The Illumination of Motion
Picture Projectors." The paper appears
in this issue of the Transactions.
About 30 members were present.
A paper on "Train Lighting" by Mr.
J. L. Minick of the engineering depart-
ment of the Pennsylvania Railroad is
scheduled for a meeting on May 9.
New Members.
The following twenty-four applicants
were elected members of the society at
a meeting of the council, April 11, 1913:
Ashley, Edward E-, Jr.
Consulting Engineer, Starrett &
Van Vleck, 45 East 17th Street,
New York, N. Y.
Austrom, Charles A.
Assistant Chief Engineer, The
Travelers Insurance Co.r Hartford,
Conn.
TRANSACTIONS I. E. S. — PART I
Barrows, Robert Y.
Chief Designer, The Mitchell Vance
Company, 507 West 24th Street,
New York, N. Y.
Burrows, W. R.
General Electric Co., 5th & Sussex
Streets, Harrison, N. J.
Coles, J. M.
Mgr. City Dept. and Engineer, Gen-
eral Gas Light Co., 80 Murray
Street, New York, N. Y.
Cox, Edward L.
Secretary, The Enos & Watkins Co.,
36 West 37th Street, New York,
N. Y.
Dibelius, Ernest F.
The New York Edison Company,
124 West 42nd Street, New York,
N. Y.
Deshon, F. B.
Deshon-Davidson Co., z22> Main
Street, Tulsa, Okla.
Dawsen, H. E.
Genera! Electric Company, Harri-
son, N. J.
Harrington, R. E.
General Electric Company, Harri-
son, N. J.
Hewlett, Arthur T.
Hewlett-Basing Studio, 298 Fulton
Street. Brooklyn, N. Y.
HlPPLE, W. C.
Superintendent, Westinghouse Lamp
Co., 514 West 23rd Street, New
York, N. Y.
Ingraham, Edgar B.
General Electric Co., 30 Church
Street, New York, N. Y.
Kerr, Thomas T.
The New York Edison Company,
124 West 42nd Street, New York,
N. Y.
LePage, Clifford B.
Stevens Institute of Technology,
Hoboken, N. J.
Mullen, Homer.
General Electric Company, Harri-
son, N. J.
Oday, A. B.
General Electric Company, Harri-
son, N. J.
Peck, Robert C.
Electrical Testing Laboratories, 80th
Street & East End Avenue, New
York, N. Y.
Perry, J. W.
General Manager, H. W. Johns-
Manville Company, Madison Avenue
& 41st Street, New York, N. Y.
Shakin, Victor.
Electrical Testing Laboratories, 80th
Street & East End Avenue, New
York, N. Y.
Summers, John A.
General Electric Company, Harri-
son, N. J.
Thistlewhite, R.
New York Electrical School, 39
West 17th Street, New York, N. Y.
Warner, J. Paul.
Iron City Engineering Co., 1172
Frick Annex Bldg., Pittsburgh, Pa.
Wyatt, Chas. K.
Salesman, H. W. Johns-Manville
Company, Madison Avenue & 41st
Street, New York, N. Y.
Sustaining Members.
At a meeting of the council, April
11, 1913, the following companies were
elected sustaining members of the
society :
Consolidated Gas Company of Boston.
The Benjamin Electric Company.
The Commonwealth Edison Company
of Chicago.
The Edison Electric Illuminating Co.
of Brooklyn.
National Electric Lamp Association.
The Macbeth-Evans Glass Company.
Westinghouse Lamp Company.
TRANSACTIONS I. E. S. — PART I
German I. E. S.
The first general meeting of the Ger-
man Illuminating Engineering Society
was held on the 25th of February in the
large auditorium of the Physical Insti-
tute of the University of Berlin.
The meeting was called to order by
the chairman of the temporary council,
Dr. E. Warburg, president of the
Physikalisch-technische Reichsanstalt. In
his address President Warburg stated
that the chief aim of the society was
to attain agreement regarding the light
unit, regarding nomenclature and re-
garding methods of measurement, and
to establish standards. He emphasized
very strongly the necessity for inter-
national agreement in these matters and
pointed out the great significance of
such agreements for the development of
science and industry. He pointed out
the function of the society in uniting
and bringing into harmony, theory and
practice as regards illumination, and
stated that on account of these impor-
tant factors, the Physikalisch-technische
Reichsanstalt had done its utmost to
further the formation of the society.
Especially interesting was a statement
that at the Reichsanstalt. experiments
are in progress from which they have
great expectations of realizing a primary
standard of light which will receive
international sanction. This is founded
on the use of the black body as a radia-
tor, held at a definite temperature.
The secretary, Dr. E. Liebenthal, re-
ported that 51 had taken part in the
meeting for organization in November
and that now the number of members
had grown to 211. The committee on
permanent officers reported, and the
following were elected : president,
Prof. Dr. E. Warburg; vice-president,
Geheimrat hummer ; vice-president. Ge-
heimrat Harber; secretary. Prof. Lieben-
thal ; secretary, Dr. Kriiss ; treasurer,
Direktor Schaller; president of the
council, Geh. Ober-Postrat Dr. Strecker;
chairman of committee on light unit,
Geh. Rat Hagen; chairman of com-
mittee on nomenclature, Dr. Strecker;
chairman of committee on methods
of measurement, Geh. Rat Brodhun.
— Tra nslated from Zeiisch rift fu r
Releuchtungswesen by C. H. Sharp.
Glare from Reflecting Surfaces.
The committee on glare from reflect-
ing surfaces is preparing an eight-page
pamphlet which will bring to the atten-
tion of the reader the necessity of
eliminating glare from glazed paper,
polished or glass desk tops, glazed
blackboards and waUs. Two leaves of
the pamphlet will be of highly calen-
dered stock, while the two remaining
leaves will be of unglazed book paper.
It is hoped that this pamphlet will im-
press upon each reader the importance
of eliminating from general use all
polished surfaces and especially glazed
paper. As paper is the most common
source of glare with which the com-
mittee is concerned, data are constantly
being collected on printing processes
and available matt-surface papers. This
enables the committee to co-operate with
publishers who desire to enlist in the
army of vision conservationists.
Factory Lighting Legislation.
Quoted below is a section of Bill
No. 26 of the laws of the State of
New York entitled "An Act to amend
the labor law, in relation to the pro-
tection of employees operating machin-
ery, dust creating machinery, and the
lighting of factories and workrooms.''
TRANSACTIONS I. E. S.— PART I
This section of the bill, which relates
particularly to lighting of factories,
passageways and workrooms, was
drafted in accordance with recommen-
dations made by the committee on fac-
tory lighting legislation of the Illumi-
nating Engineering Society. The bill
was signed by Governor Sulzer April
i/. IQI3-
All passageways and other portions of a factory,
and all moving parts of machinery which are not
so guarded as to prevent accidents, where, on or
about which persons work or pass or may have to
work or pass in emergencies, shall be kept prop-
erly and sufficiently lighted during working
hours. The halls and stairs leading to the work-
rooms shall be properly and adequately lighted,
and a proper and adequate light shall be kept
burning by the owner or lessee in the public
hallways near the stairs, upon the entrance door
and upon the other floors on every work day in
the year, from the time when the building is
open for use in the morning until the time it is
closed in the evening, except at times when the
influx of natural light shall make artificial light
unnecessary. Such lights shall be so arranged as
to insure their reliable operation when through
accident or other cause the regular factory light-
ing is extinguished.
All workrooms shall be properly and adequate-
ly lighted during working hours. Artificial il-
luminants in every workroom shall be installed,
arranged and used so that the light furnished
will at all times be sufficient and adequate for
work carried on therein, and so as to prevent
unnecessary strain on the vision or glare in the
eyes of the workers. The industrial board may
make rules and regulations to provide for ade-
quate and sufficient natural and artificial light-
ing facilities in all factories.
This act shall take effect October first, nine-
teen hundred and thirteen.
0-
TRANSACTIONS
OF THE
Illuminating
Engineering Society
APRIL, 1913
PART II
Papers, Discussions and Reports
[ APRIL, 1913 ]
CONTENTS -- PART II
A Photometer Screen for Use in Tests of Street Illumina-
tion. By Arthur H. Ford 155
The Flame Carbon Arc Lamp. By W. A. Darrah 162
The Illumination of Motion Picture Projectors. By J.
Frank Martin 180
/&V
A PHOTOMETER SCREEN FOR USE IN TESTS OF
STREET ILLUMINATION.
BY ARTHUR H. FORD.
Synopsis: This paper puts forth the proposal that in illumination
measurements a test-plate with rounded surfaces or one with several plane
surfaces, the mean illumination of which has been determined, be used
in place of the usual flat translucent photometer test-plate. It gives
comparative illumination readings obtained with the use of five different
test-plates.
The purpose of street illumination being the making visible
of obstructions to traffic, which obstructions seldom have plane
surfaces making definite angles with the street surface, the
ordinary method of measuring street illumination on a horizontal
plane or one normal to the ray of light does not give the infor-
mation desired. The writer proposes to overcome this defect
by using, as a screen, a body with rounded surfaces or several
plane surfaces; the mean illumination of which is determined.
Such a screen would correspond to a brick if plane surfaces were
used or a stone if rounded surfaces were used ; and the illumina-
tion would be, to a considerable extent, independent of the
direction from which the light comes.
This paper is the record of some tests with screens having
various configurations.
Since the surroundings of the photometer cannot be controlled
in street photometry as in laboratory work, it is obvious that the
surfaces used must be those of a translucent body attached to
the viewing part of the photometer. The screens tested were
mounted, in turn, on the elbow tube of a Sharp-Millar pho-
tometer and readings of the instrument 'made as the angular
position of the test lamp was changed through 90 deg. Polar
curves were then plotted between the angular position of the
test lamp and readings of the photometer, in terms of the
maximum reading obtained with the particular screen being
tested.
i56
TRANSACTIONS I. E. S. — PART II
The following screens were used:
No. i. — A plate of translucent, milk white glass with a ground
surface (Fig. i), the glass being the regular diffusing screen fur-
nished with the photometer.
No. 2. — A cube of paraffin i^ inch (4.1 cm.) on a side; one
side covered with an opaque screen (Fig. 2). While paraffin
is not a suitable substance for making permanent screens, the
ease of moulding it into irregular shapes adapts it admirably for
making screens for temporary use. Its optical properties are
satisfactory if the screens are carefully selected for uniformity
of texture and optical symmetry.
No. 3. — One quarter of a paraffin sphere 1^ inch (4.1 cm.)
in radius having one flat side covered with an opaque screen
(Fig. 3)-
No. 4. — An "Alba" glass hemisphere 3 inches (7.6 cm.) in
diameter having one half blackened Fig. 4). This globe is about
yi inch (0.3 cm.) thick; the translucence being due- to air bubbles
in the glass.
No. 4a. — The same screen as No. 4, but with the addition of
a piece of paper on the flat side of the hemisphere.
The following data were obtained:
Screen No. i.
Angle of
lamp.
Photometer
reading.
90
80
IOO
97
70
60
90
81
50
40
30
70
60
46
20
34 .
IO
—
O
—
Screen N° I.
ford: a photometer SCREEN
157
Screen No. 2.
Angle of
lamp.
90
80
70
60
50
40
30
20
IO
O
Angle of
lamp.
90
80
70
60
50
40
30
20
IO
o
Angle of
lamp.
90
70
60
50
40
30
20
10
o
Photometer
reading.
57
65
81
85
100
100
95
96
99
75
®IOO
Photometer
reading.
75
87
94
97
100
100
93
9i
84
66
Screen No. 3
°\0°
Screen No. 4.
Screen Ns 3.
Fig- 3-
Photometer
reading.
IOO
70
52
46
42
37
35
33
28
22
158
TRANSACTIONS I. E. S. — PART II
Screen No. 4a.
Angle of
lamp.
90
80
70
60
50
40
SO
20
IO
o
°»0° AftO
Photometer
reading.
70
82
92
IOO
IOO
97
9i
87
72
61
Screen N°4A.
Fig. 4a.
CONCLUSION.
Screens Nos. 3 and 4a are satisfactory for use in street pho-
tometry, with the advantage somewhat in favor of No. 3 because
of the small dependence of illumination determinations on the
direction of the light source. The writer would suggest the use
of a screen of this form made of "Alba" or some similar glass.
DISCUSSION.
Dr. C. H. Sharp (communicated) : Prof. Ford makes the
statement that the ordinary method of measuring street illumina-
tion on a horizontal plane or on one normal to the ray of light,
does not give all the information requisite to determine the
visibility of objects on the surface of the street. This is quite
true. He proposes to overcome this difficulty by using a test screen
which gives a value which represents an indeterminate mixture of
illumination values in various planes. It is not at all clear that
this represents a solution of the difficulty.
In the first place, it must be noted, as Millar has pointed out,
that objects on the street are seen because of a difference of the
brightness of the object and its background. A brick lying on the
street may be seen because it is whiter than the street surface ; or
because to the observer a vertical surface of the brick is brighter
than the horizontal surface of the pavement, a condition which
may be due either to the brick being actually whiter or to the
A PHOTOMETER SCREEN 159
illumination on the vertical plane being greater than that on the
horizontal plane ; or it may be seen because a vertical surface of
the brick is darker than the street surface, which may be because
it lies in its own shadow. There are then various conditions
which determine visibility and which arise from the relations be-
tween the object viewed, the lamps illuminating it, and the rela-
tive reflecting power of the object and of the street surface, as
well as upon the illumination in various planes. Therefore, even
if we know all there is to be known about the illumination in all
the various planes conceivable, we could not say that any brick
would be visible from any point unless we had some other items
of information besides. Illumination values alone would not tell
us. Therefore at best Prof. Ford's solution is an imperfect one.
From a scientific point of view also his proposition seems to
be untenable. When we consider the illumination at a certain
point on a certain plane, we are dealing with a perfectly definite
physical quantity. The only arbitrary question is how perfectly
the test plate used in making the measurement of that illumination
conforms to the ideal law of diffusion. When, however, we
measure the brightness of a certain piece of parafhne or a certain
hemisphere of a certain kind of glass, we have a result which is
purely arbitrary and incapable of interpretation in simple terms.
It is also very questionable whether hemispheres of diffusing glass
could under the best conditions be made so like each other that
they would always show the same diffusion characteristics.
Unless it were possible to reproduce these hemispheres, it would
be impossible to make all illumination photometers give the same
results under the same conditions. And there is no reason to
suppose that in the present state of the art of glass making, this
can be done.
It would be interesting if Prof. Ford were to study the illumina-
tion on a street using his proposed test plate, and were then to
present a paper to the society telling just what that illumination
was in actual physical units.
Mr. M. Luckiesh (communicated) : Prof. Ford presents
some interesting data regarding the problem of measuring
"illuminating efficiency." It is true that street lighting presents
some distinct problems in the matter of rating illumination
2
160 transactions i. e. s. — part ii
according to its ability to make objects visible but interior lighting
is not free from the same problems. Various methods for
measuring "illuminating efficiency" have been suggested but the
one used at present, although deficient, will no doubt be adhered
to until illuminating engineers agree on the answer to the ques-
tion. How does the direction of the incident light affect the
ability of that light to make objects visible?
This question will not be answered without fully analyzing
illumination. With this in mind the writer has studied in detail
the distribution of natural and artificial light in interiors. This
work which appeared in the Transactions for October, 1912, is
only a beginning of the analysis which must be made in order to
conclude just how lumens incident from various directions should
be weighted.
Prof. Ford's screens, Nos. 3 and 4a, which he considers satis-
factory for use in street photometry take into account light
incident within an angular range of 90 deg. (considering only the
vertical plane parallel to the direction of the street). Obviously
the screen or the whole photometer must be rotated 180 deg. in
order to account for the light from the other light source because
there would be positions where sources on opposite sides of the
photometer would contribute light. Just how he would weight
these two measurements is not stated.
To fully emphasize the complexity of the problem, consider a
case where one sees a vehicle in silhouette against a bright back-
ground such as a highly illuminated pavement. In such cases
visibility depends upon the contrast of the dark background;
therefore the less light that is incident on the vehicle from the
general direction of the observer the more visible will the
object be.
These points are cited merely to emphasize the fact that more
work must be done on the analysis of illumination. In view of
the lack of agreement regarding the method of determining
"illuminating efficiency", the writer cannot wholly agree with
Prof. Ford's conclusion that screens Nos. 3 and 4a are satis-
factory for street illumination. It would be interesting to learn
why he reached this conclusion. Prof. Ford's data are very in-
teresting and certain of those screens may perhaps find immediate
A PHOTOMETER SCREEN l6l
application in special cases. However, the data will be of greater
use after illuminating engineers reach an agreement and decide
what they desire to measure.
Mr. G. H. Sticknev (communicated) : The failure of the
normal intensity or that on a horizontal surface to express the
true values of street illumination has long been evident and not
infrequently has been the subject of discussion among illuminat-
ing engineers. While I believe it is generally conceded that the
true values lie somewhere between those determined by the above
mentioned methods, there seems to be no general agreement as to
just where these values fall, and it is quite probable that, under
different conditions, their relative position between these two
extremes may vary.
About 1906 Mr. \Y. D'A. Ryan and myself experimented a
little with a translucent hemisphere and also with other spherical
sections partially blackened or coated with tin foil, but, on
account of various difficulties, did not arrive at a form of screen
which we were willing to recommend.
Beyond the difficulty of determining the intermediate values
which would be acceptable as fair to both large and small units
with wide and narrow spacing, we were unable to secure a screen
which would make accurate measurements on very low intensities.
It will be remembered, in this connection, that the class of
problems where this type of measurement is important are those
in which the actual normal intensities may run as low as 1/100
foot-candle or less. While we obtained accurate settings with
diffusing screens of opal glass and paraffine with values above one
foot-candle, we did not have much success with low intensities;
that is, 1/10 foot-candle or less.
In looking over the curves presented by the author I would
agree that screens No. 3 and 4-a show a better characteristic
curve than the other forms. Still I believe that many observers
will feel that both of these types give too much weight to light in
angles approximating zero degrees.
\\ hile at present the difficulties seem almost unsurmountable, it
must be recognized that there is an insistent demand for a better
method of determining the value of street lighting intensities and
any work tending toward this end will be of great value.
l62 TRANSACTIONS I. E. S. — PART II
THE FLAME CARBON ARC LAMP.*
BY W. A. DARRAH.
Synopsis: The operating characteristics of the recently developed
flame carbon arc lamp upon direct and upon alternating current supply
are outlined and illustrated in the following paper. Flame carbons and
their light giving properties and characteristics are also discussed. The
author states that this lamp is peculiarly well adapted to street and large
area lighting on account of the color (which approximates a daylight
value when white flame carbons are used) and distribution of its illumi-
nation, and on account of its comparatively low operating and main-
tenance costs. The efficiency of this type of lamp (with white flame
carbons) quoted in mean lower hemispherical candles per watt is said to
vary from 2.5 with light opal globes to 5 with clear glass globes ; yellow
carbons, the author adds, may be obtained which will give from 30 to
50 per cent, more light for the same energy consumption.
The methods of lighting to-day are being revolutionized.
Hundreds of thousands of dollars worth of old equipment, is
yearly being sent to the scrap pile, to be replaced by more
efficient, more effective, more economical apparatus. The num-
ber of "great white ways" is being multiplied monthly. Civic
bodies and merchant's associations are clamoring for more light.
One of the fundamental causes of the transition now in pro-
gress, is the rapid, and very complete development and com-
mercialization of the long-burning, flame carbon arc lamp. The
flame carbon arc lamp is not new. It does not even in its present
highly developed form, depend for its operation upon new
theories, or recent discoveries. But it does depend for its present
swift and complete development upon the awakening of civic
pride, the discovery of the enormous commercial value of light,
and the accompanying prosperity which has swept across the
country.
The subject of long burning flame carbon arc lamps is too large
a one to be treated in a limited paper of this kind. It is proposed
here to very briefly touch -on some characteristics of long burning
flame carbon arc lamps, to show how these characteristics particu-
larly adapt this illuminant for some fields of work, and to then
*A paper read before a meeting of the New England section of the Illuminating
Engineering Society, February 17, 1913.
DARRAH : THE FLAME CARBON ARC LAMP 1 63
consider some points in connection with the design of these lamps
from the standpoint of the operating man.
The long burning flame carbon arc lamp, is to-day, without
doubt, the most efficient, commercial source of light considering
at the same time color value; and what is unfortunately not
always true of illuminants of this class, it is also the most
economical to operate, considering the average conditions of cost
of energy, labor, and materials when the mean lower hemispheri-
cal candle-power required exceeds five or six hundred.
The light of the flame carbon lamp is steady and the dis-
tribution excellent for lighting large areas with considerable
uniformity. The color may be varied within a rather wide range
from white to yellow depending upon the kind of carbons em-
ployed. Other colors such as red, blue, green, etc., may of course
be obtained if sufficient demand for them should arise, as, for
instance, for advertising purposes or where special colors lend
themselves more readily for special work.
A further highly desirable attribute of the flame arc, and one
which has perhaps not yet been fully appreciated is the reduced
intrinsic brilliancy over the carbon arc, due to the relatively large
area of the light source. While the intensity of the flame arc is
so great that even with the much larger area of the light source
it is usually desirable to use diffusing glassware, yet the shadows
are far softer and the distribution more uniform than can be
obtained with the older forms of arcs.
CHARACTERISTICS OF THE FLAME CARBON ARC.
A consideration of the flame carbon arc is only of value in con-
nection with a lamp and a specific variety of carbons. In the
statements which follow an enclosed flame carbon arc lamp
similar to that shown in Fig. 1 is considered, although the dis-
cussion in general applies to other lamps of this type. The car-
bons under consideration are assumed to be commercial carbons,
varying in diameter from V$ in. to % in.
In appearance, the flame carbon arc more nearly resembles the
metallic flame arc than any other type. It consists of a long ill-
defined flame which is intensely luminous and is terminated with
a bright point at each end. Both the alternating and direct cur-
164 TRANSACTIONS I. E. S. — PART II
rent arcs have these characteristics, and it is very difficult to dis-
tinguish the direct current from the alternating current arc by
inspection. No crater is formed on the positive carbon and the
distinct cathode spot on the negative carbon is less conspicuous.
It is probable that the length and density of the arc material
prevents the scouring action of the particles emitted by the arc,
which probably cause the crater in the case of the open direct cur-
rent arc. Unlike the enclosed carbon arc, the flame is intensely
luminous and is the source of a considerable amount of radiation
in the red and green portions of the spectrum.
The majority of light comes from the luminous flame, although
the terminals of the arc have a higher intrinsic brilliancy than the
other portions.
The arc may have any color, depending upon the materials used
in the carbons. Commercially, white and yellow light carbons
are mainly employed, although under special conditions, for
special purposes, other colors are occasionally used. The color
of the arc is dependent upon the current density to some extent.
In other words, an arc that would be white with certain values of
current density, may become distinctly yellow when the current
density is materially decreased. It appears that the temperature
of the arc also has considerable effect upon the color, since an arc
which will burn white under normal conditions, may become in-
tensely yellow when the temperature is lowered, either by cooling
the terminals, by special air draft, or by forming the arc in an
atmosphere which will readily conduct the heat away. It seems
probable that the change in color may be produced by lowering
the temperature of the minute particles which are luminous, to
a yellow heat instead of allowing them to remain at a tempera-
ture sufficiently high for the radiation of white light.
It is interesting to note that the addition of water vapor to the
arc chamber will cause an arc normally emitting a white light to
burn yellow. A further suggestive fact is that certain gases may
be present in the arc chamber which will cause a yellow arc to*
become intensely white.
Fig. 2 shows an alternating current flame carbon arc between
Y$ in. carbons and an atmosphere of carbon monoxid and nitrogen.
This was a 10-ampere arc with a potential between the electrodes;
<r
ih
Fig. i.— Enclosed flame carbon arc lamp.
Fig. 2. — A io-ampere, 50-volt alternating current flame carbon arc between % in.
carbon and an atmosphere of carbon monoxid and nitrogen. The
magnetic blow away from the side rod is clearly shown.
-A 6.5-ampere, 70-volt direct current flame arc between % in.
carbons under normal conditions.
DARRAH : THE FLAME CARBON ARC LAMP 165
of approximately 50 volts. In this arc, the magnetic blow away
from the side rod may be distinctly noted.
Figs. 3 and 4 show direct current, 6.5-ampere, 70-volt arcs be-
tween % in. carbons. It will be noted that both ends of the
direct current arc are distinctly luminous and very similar in
appearance to the alternating arc. Fig. 3 shows the arc under
normal conditions, while Fig. 4 shows the so-called non-magnetic
arc which is considerably diffused and very slightly affected by a
magnetic field. Figs. 5 and 6 illustrate an enclosed carbon arc
between unimpregnated y2 in. carbons operating under similar
conditions to the flame carbon arc. It will be noted here that the
electrodes of the enclosed carbon arc are intensely bright. Since
the arc between the unimpregnated carbons emits a large per-
centage of blue and ultra-violet rays, the photograph shows this
arc considerably more luminous than it appears to the eye.
The diameter of the flame carbon arc under normal conditions
is approximately Y% in. Curve 1, of Fig. 7, shows the length of
the flame carbon arc with various voltages applied to the
terminals and with constant current. It will be noted that the
length of the arc ranges from )A in. with 40 volts, to approxi-
mately 5 in. with 220 volts. The increase is substantially a
straight line. If continued, the curve would pass through the zero
separation point with about 30 volts drop, indicating that this is
the loss at the terminals independent of the length of the arc.
Calculated on this basis, an arc 5 in. in length and having a total
drop of 220 volts would have 190 volts loss throughout its length,
or a drop of 38 volts per inch. This is equivalent to about 3.8
ohms per inch of length for a 10-ampere, direct current arc, which
while somewhat higher, is not very materially different from the
conditions often obtained in the enclosed carbon arc.
Curve 2, on Fig. 7, is the well-known characteristic curve of
an arc. It will be noted that when the current exceeds approxi-
mately 6.5 amperes, the flame carbon arc' becomes exceedingly
stable. In other words, the drop in voltage across the arc in-
creases in approximately the same ratio that the current is in-
creased. This is somewhat different from the titanium and
enclosed arcs which are correspondingly very much more un-
stable. The causes of the stability of the flame carbon arc on
l66 TRANSACTIONS I. E. S. — PART II
currents above 6.5 amperes are mainly two : first, the relatively
high resistance of the arc and the increase of this resistance as
the length increases ; and, second, the lower evaporation point and
greater volume of the materials in the flame carbon arc. In this
connection, it is interesting to note that these two conditions, to-
gether with the higher margin between the temperature at which
the arc is a conductor and the tempearture at which it is not, allow
of the satisfactory operation of the flame carbon arc upon cir-
cuits, the frequencies of which is as low as 25 cycles.
Figs. 8, 9 and 10 respectively, show oscillograms of the
titanium arc, the enclosed carbon arc and the flame carbon arc.
It will be noted that while the current wave has substantially the
same distortion in the case of the titanium and enclosed carbon
arcs, it is very much more nearly a sine wave in the case of the
flame arc. An inspection of the oscillographs shows that the dis-
tortion of the voltage wave is very considerably greater than the
distortion of the current wave. The titanium arc shows a dis-
tinct peak at the instant at which the current is interrupted and a
very steep wave front. This distortion of the current and volt-
age wave in the titanium arc is so great and their centers are
displaced to such a degree that the resultant power factor is
approximately 50 per cent.
The distortion of the enclosed carbon arc is somewhat less, but
the displacement of the current and voltage waves is sufficient
to reduce the power factor to approximately 80 per cent. It is
interesting to note the higher harmonics which are present in
the voltage wave of the enclosed carbon arc are absent in both
of the flame arcs.
The voltage wave of the flame carbon arc shows less distortion
than either of the other arcs, and the displacement is so slight
that the power factor ranges from 85 to 90 per cent. The above
discussion should be understood to apply entirely to the power
factor of the arc itself and to be independent of the lamp
mechanism or coils.
The advantages of the higher power factor are well known,
while the advantages of the smoother wave form are : lower
voltage strains, more stable arcs, and less induction between arc
circuits and adjacent telephone or telegraph circuits.
4- — A 6.5 ampere 70-volt direct current flame arc between % in. carbons. The
so-called non-magnetic arc which is said to be only slightly affected by a
magnetic field, and considerably diffused, is plainly illustrated.
F»g- 5.— An enclosed carbon arc between
' : inch uniinpregnated carbons.
Fig. 6. — An enclosed
carbon arc.
Fig. 7. —Characteristics of flame carbon arc lamps.
Fig. S. — Oscillograms of titanium arc.
Fig. 9. — Oscillograms of enclosed carbon arc.
DARRAII : THE FLAME CARBON ARC LAMP
l67
As previously pointed out, the smaller distortion and higher
power factor of the flame carbon arc results largely from the
sloping characteristic curve which is due to the greater margin
between the temperature at which the gases of the arc conduct
and the temperature at which they do not, to the greater volume
of volatile material in the arc and the lower vaporization point
of this material.
The flame carbon arc is inherently a large energy light unit.
Since the light is emitted quite largely by the flame, it is desirable
to have the flame as long as possible. Since the drop at the
electrodes is fixed at approximately 30 volts and since the resist-
ance drop of the flame is approximately 38 volts per inch (for
a 10 ampere arc) it will be evident that a comparatively high
Fig. 10.— Oscillograms of a flame carbon arc.
voltage is desirable for most economical operation. This state-
ment, together with all other statements regarding the operating
characteristics of the arc, should be considered as applying to
the commercial type of carbons at present on the market, and it
should be kept in mind that it is possible to materially vary these
operating characteristics by employing special types of carbons.
The present current densities which have been found most
satisfactory are as follows :
Per square inch.
100 amps.
{1) Current density at the arc 100
(2) Current density at the arc, negative terminal, ap-
proximately 1 ,500
(3) Current density at the arc, positive terminal, with
diffused arc - 35
\\ ith commercial carbons, best results are secured with approx-
l68 TRANSACTIONS I. E. S. — PART II
imately 10 amperes through the arc and it is upon this basis
that the above statements are made. Since the power factor of
the flame carbon arc is approximately 85 per cent., it will be
evident that the wattage under which most satisfactory operation
may be obtained lies between 350 and 500 watts, more economical
operation being obtained with a higher wattage.
CONSIDERATION OF FLAME CARBONS.
As the carbons mark one of the greatest differences between the
present flame carbon arc lamps and the old solid carbon arc
lamps, some space may properly be devoted to their consideration.
The flame carbon differs from the solid carbon in that certain
chemicals are added during their manufacture with the object
of increasing the light; securing better operation, and reducing
slag troubles. The chemicals commonly added may be classified
as follows :
( 1 ) Illuminants. — These comprise the compounds of three or
four peculiar elements. For white light, cerium and titanium
offer the greatest possibilities ; while calcium and, to some extent,
tungsten form the main illuminants in yellow light carbons.
(2) Sustainers. — Since the flame carbon arc, as illustrated in
Figs. 2, 3 and 4 is very considerably longer than the enclosed
carbon arc, it would be prohibitively unstable unless certain com-
pounds were added to remedy this defect. At present, fluorides
and to a small extent, borates, form the most suitable sustainers
when slag troubles, evaporation point and cost are considered.
These elements are frequently introduced as compounds of the
illuminants.
(3) Conductors. — In order to overcome certain troubles which
arise due to slag which may form in case the carbon is consumed
more rapidly than the illuminants and sustainers, certain com-
pounds are frequently added to prevent slag troubles which
might otherwise result.
By a series of long experiments — for theory does not seem to
be at present sufficiently developed to do much more than indi-
cate the direction of progress — the large carbon manufacturers
have to-day developed flame carbons which are entirely commer-
cial and which give excellent results, both from the standpoint
of efficiency and operation.
darrah: the flame carbon arc lamp 169
In view of the scarcity of consistent data, it is perhaps, dan-
gerous to venture a theory to account for the high efficiency of
the flame carbon arc. One explanation, however, which seems
reasonable and which accounts for a large number of facts which
have been observed, is to consider the arc as a place where the
mineral compounds with which the carbons are impregnated are
raised to such a temperature that dissociation is continually in
progress and that at very short distances from these points of
high temperature, re-combination is occurring very vigorously.
In this process of dissociation and re-combination, the elements
which form the illuminants become raised to the temperature
necessary for selective radiation.
A consideration of the position in Mendeleff's table of the
elements which form the illuminants of the arc. and of the
chemical characteristics of these elements, makes the above theory
appear more probable. In addition to the high temperature
necessary to allow selective radiation, it is essential that the
illuminant be composed of a material which readily forms com-
pounds having a very high melting point and vaporization point.
All of these requirements seem to be met by the oxides of cerium,
calcium and tungsten. In this connection — as previously pointed
out — the effect of cooling the arc and thereby changing its color
from white to yellow is very suggestive.
Fig. 10a shows the variation of candle-power which occurs
with increased voltage on a 10-ampere. alternating current flame
carbon arc between various varieties of carbons. These curves
are interesting as indicating that, with few exceptions, the candle-
power increases directly with the voltage until a certain value is
reached, and in some cases somewhat more rapidly. This is to
be expected, as the increased voltage allows an increased arc
length and therefore a longer flame. When a certain value of
voltage and a certain arc length has been exceeded, additional cool-
ing effects introduced prevent a material increase in the total light
flux emitted, and for this reason, the curve flattens out after pass-
ing a certain critical point.
Fig. 1 1 shows the effect of increased currents on a 48-volt,
60-cycle arc under normal conditions. It will be noted that the light
emitted increase uniformly with the current over the majority of
170
TRANSACTIONS I. B. S. — PART II
the range given. Depending upon the amount of illuminants in
the current increases. This affords a rough means of determining
the amount of illuminant in a given grade of carbons. In this
connection, it is interesting to note that the addition of illuminants
to a flame carbon does not increase the total light emitted in a
direct ratio to the amount of illuminants added. In other words,
as the percentage of cerium oxid in a white-light carbon is in-
creased, the candle-power first increases until a maximum value
Fig. 10a. — Variation in candle-power with increase of voltage across arc
of flame carbons.
is reached, after which the addition of more cerium oxid will
decrease the intensity of the light emitted for a given amount of
electrical energy. This "condition together with the effect of slag,
which may be very serious in improperly designed carbons, limits
the amount of mineral material which may be added to the
carbons.
The difficulties introduced by slagging of the flame carbons has
DARRAH : THE FLAME CARBON ARC LAMP
171
received considerable attention and has been overcome com-
mercially. Assuming that these carbons are properly designed,
the outage during normal, commercial operating conditions should
not exceed y2 to $4 of 1 per cent., provided the lamps are main-
tained in the proper condition.
Slagging may be caused by the entrance of air into the arc
chamber, due to imperfect globe seats or leaking condensers. The
excess of air causes the consumption of the carbons to proceed
more rapidly than the mineral components can be vaporized.
Kig. 11. — Variation of candle-power with increase of current of arc
with flame carbons.
This allows an accumulation of fused oxids and fluorids upon the
surfaces of the electrodes, thus forming when cold, an insulating
layer. An excessively low current or an arc longer than normal
may also cause slagging.
None of the difficulties mentioned above should be encountered
in the commercial lamps providing proper globes are used and the
lamps are maintained in good condition. A very small amount
of slag is no detriment to the operation of the flame carbon lamp
172 TRANSACTIONS I. E. S. — PART II
since the hammer blow which is given when the carbons fall to-
gether prevents trouble of this nature from being serious.
From a mechanical standpoint, it may be stated that the car-
bons now commercially obtainable are entirely satisfactory.
Present diameters vary from approximately 24 m- to ]4 in->
depending upon the conditions under which they are used. It is
possible to secure these carbons within 0.025 in. from the
specified diameter without adding appreciably to their cost. Since
this is true, it will be evident that a ring clutch may be employed
in the operation of flame carbon lamps with entire satisfaction.
Since the average flame carbon contains more than 30 per cent,
of solid material which is not consumed, but which must be
vaporized by the arc, the problem of disposing of this material
naturally required solution. It was found that this solid material
(called "soot" by lamp operators) would not condense upon sur-
faces which were maintaind at an elevated temperature, while it
will readily condense in the form of a soft white powder upon
any cool or relatively cool surface. This is taken advantage of
by so arranging the globes of the lamps that they will be ma-
terially hotter than a second and communicating chamber into
which the gases from the arc are forced.
The second chamber, which is made of metal and designed to
expose a comparatively large surface to the air, is called the "con-
densing chamber" (Fig. 12), shows an outline sketch of a flame
carbon lamp and indicates the path of the gases from the arc. It
will be noted that this lamp is so designed that the majority of
the gases which carry the vaporized mineral material pass directly,
into the condensing chamber where the majority of the soot is
deposited. The condensing chamber also contains sticks of
magnesia or other alkaline material which will readily combine
with the hydrofluoric acid, nitric acid, etc., preventing the free
acids from attacking the globe, thereby decreasing its transparency
and the amount of light emitted.
An examination of the arc through an absorbent glass shows
a very rapid movement of the gases immediately surrounding the
arc. Streams of semi-luminous vapors may be noted being pro-
jected violently upward from the arc flame. These vapors con-
DARRAH : THE FLAME CARBON ARC LAMP
173
tain the so-called "soot," together with hydrofluoric, nitric and
sulphuric acids.
Since the temperature of the arc in its cooler portions prob-
ably exceeds 2.500 deg. C, while the remainder of the arc
chamber is filled with gas at a temperature not exceeding 150
deg. C, it is obvious that the different density of the two gases
is sufficient to account for the violent motion in the vicinity of
the arc.
In connection with the wash of air from the arc around the
upper carbon is found an explanation of the markedly different
Fig. 12. — Sketch indicating path of gases from flame carbon arc in one type of lamp.
rates of consumption of the upper and lower carbons. In other
words, the surface of the upper carbon over which the hot gases
pass is very rapidly consumed by any free oxygen which may be
in the arc chamber.
The life of a flame carbon depends upon its density, the cur-
rent, the length of the arc, the composition, and above all, the
tightness of the enclosure in which the -carbon is consumed.
Every effort should be made to maintain the enclosure as perfect
as possible at all times. Only accurately ground globes, free
from flaws at the globe seat, should be used ; while the portion
of the lamp which comes in contact with the globe should be
made from a hard, strong alloy which will be affected as little
174
TRANSACTIONS I. E. S. — PART II
as possible by the fumes from the arc. To insure maximum life,
it is necessary that all joints in the condensing chamber be
maintained tight and that the condenser be removed as little as
possible. Under commercial conditions, a life of from 15 to
20 hours per inch may be obtained from standard carbons.
LAMP PERFORMANCE.
Passing from the somewhat more theoretical considerations of
the arc to the more practical point of operation and performance,
Distribution Curve of Clutch Type Fume- Carbon Arc Lamps
Clear Inner and A/bo Outer Globes. White Carbons.
Terminal Volts £2
Terminal Watts 425
/Ire Volts . ... 46
Arc Watts 395
Terminal Volts J/S
Terminal Walts 747. S
Arc Volts 70
Arc lA/atls 455
A. C. S£f?/£S
Amperes 10
EfT/c/ency...:..93X
Power Factor 816%
D.C. Multiple
Amperes 6.5
Efficiency.. ..6l/i
M.L.H.C.P..... .1000
M.S.C.P. 656
M.LUCP Permit 235
M.5.CP ■■ ■■ 1.54
M.l.H.CP. 820
M.S.CP. 500
M.LH.C P Per Wolt 1.09
MS.CP. ■ "6?
Fig. 13.— Comparative distribution curves of a direct current, 6.5 ampere multiple
and a 10-ampere series alternating current flame carbon arc lamp.
the distribution curve of the flame arc deserves some consider-
ation.
Fig. 13 shows comparative distribution curves of the direct
current, 6.5-ampere, multiple lamp and the alternating current,
10-ampere, series lamp. It will be noted that the two distribu-
tion curves are very similar. The slight difference in shape
between the distribution curve of the alternating current and
the direct current lamps is due, mainly, to the long arc of the
DARRAH : THE FLAME CARBON ARC LAMP 175
direct current lamp and to the somewhat greater intrinsic bril-
liancy of the lower portion of the direct current arc.
It will be noted that these distribution curves indicate that a
very considerable amount of light is radiated at angles above
45 deg., the maximum illumination occurring at approximately
30 deg. below the horizontal. For street lighting or for the
illumination of large areas, this is a very desirable characteristic,
and in comparing various lamps of the same general types, it is
often desirable to rate the lamps on such a basis that the candle-
power between the horizontal and 30 deg. below the horizontal
will be given more weight than the candle-power at other angles.
In other words, the total useful light flux represented by the
candle-power at any given angle varies with the cube of the
cosine of the angle.
From an esthetic standpoint, the flame arc is particularly desir-
able because of the relatively low intrinsic brilliancy of the light
source and the comparatively large area from which the light is
emitted ; and this makes an illumination from the arc softer and
less fatiguing to the eye. It also avoids to a large extent the
harsh shadows which are so prominent with the enclosed carbon
arc. In spite of the lower intrinsic brilliancy of the flame carbon
arc, it is desirable to employ the diffusing glassware.
The distribution of energy in the spectrum of the flame carbon
arc is peculiarly well suited for illumination which is intended to
approximate daylight, although where the matching of color
values is of special importance the carbons employed should be
designed with this point in view. In the light from white car-
Dons, the red and green wave lengths are found in very materi-
ally greater proportions than in the enclosed carbon arc or
metallic flame arcs.
The efficiency of the flame arc as an illuminant is particularly
high. The mean lower hemispherical candle-power per watt
varies from 2.5 with white carbons and "Alba" glassware, to
5 mean lower hemispherical candle-power per watt with clear
glassware. When opalescent globes are employed, the efficiency
falls between these values. Yellow carbons may be obtained
which give from 30 to 50 per cent, more light than the white
carbons.
3
I76 TRANSACTIONS I. E. S. — PART II
From an operating and maintenance standpoint, the flame
carbon lamp is peculiarly suited for the illumination of city
streets, as well as for factory lighting. The lamps require but
little attention ; the life between trims is approximately 100
hours, and the time and labor of trimming the lamps is small.
Under efficient direction, one man could very readily trim and
clean from 50 to 75 lamps per day. At each trimming period,,
it is desirable to carefully remove the "soot" from the globes
and condensing chamber, while after each 1,000 hours of opera-
tion, the consumer should be renewed. Depending upon the
care and efficiency with which the installation is maintained, the
life of the globes varies from 2,000 to 5,000 hours. The impor-
tance of maintaining the globes clean and in good condition is
frequently not sufficiently appreciated by operating engineers,,
and as a result, the light efficiency of the lamp is very materially
reduced.
CONCLUSIONS.
A consideration of the points noted above will make evident
some of the features which have been incorporated in modern
flame carbon arc lamps and some of the difficulties which have
been encountered and commercially overcome. At the present
time, the flame carbon arc lamp is a commercial success and has
stood the test of actual operating conditions. The field of the
flame carbon arc lamps is continually expanding, and due to
the increased prosperity of the period and the fact that engi-
neering development has kept pace with commercial require-
ments, there seems to be little doubt but that the installation of
flame carbon lamps will in the near future considerably out-
distance the installation of all other types of large energy light-
ing units, including those used in the most utilitarian and busiest
shops and factories, as well as those employed for highly orna-
mental street or parkway lighting.
THE FLAME CARBON ARC LAMP 1/7
DISCUSSION.
Messrs. W. R. Mott, R. B. Chillas Jr., and A. T. Baldwin
(communicated) : Several points in Mr. Darrah's excellent
paper bring up questions which we desire to discuss briefly.
Along with the other differences mentioned between the flame
arc and the magnetite arc it may be well to bring attention again
to the fact that the nature of the positive flame electrode largely
determines the color and candle-power of the arc, while with the
magnetite arc the negative is the determining factor.
An examination of oscillograms of flame arcs will show that
the less-than-unity power factor is the result of a distortion rather
than a displacement of the waves. The high power-factor of the
flame arc is due to the ease with which the current is re-established
through the hot gases after each reversal of current. Since the
light flux is largely dependent upon the current it is evident that
the least variation or flicker in the light will be found in the arc
of highest power factor. As indicated in the paper the starting
voltage of the flame arc is very low. In fact it is much lower for
the flame arc than for either the enclosed arc or magnetite. In
these respects the flame arc in general has a marked advantage
over the other arc illuminants. Certain types of commercial car-
bons possess these properties to a far greater degree than others.
Such carbons are particularly useful on low frequency (25 cycle)
circuits, and in lighting shops containing moving machinery,
especially if the movement of machine parts and the cyclic varia-
tion of the light are nearly synchronous.
The alternating current flame arc possesses one rather striking
advantage over the plain carbon arc because it shows less flicker
at lower frequencies. In the plain carbon arc the electrode which
is positive at any instant is giving out most of the light. This
means that on 25 cycles the light in one direction passes through
a maximum 25 times per second. In the flame arc, where the
light comes from the arc stream itself, the corresponding figures
are 50 times per second, or far beyond the critical flicker
frequency of the eye.
The best arc voltage to be used is determined to a large extent
by other factors taken in conjunction with those pointed out by
Mr. Darrah. As the arc voltage and consequently the arc length
I78 TRANSACTIONS I. £. S. — PART II
is increased at constant current a point is reached at which the
candle-power increase is not as fast as the voltage increase due to
the greater cooling of the arc and the steadiness begins appreci-
ably to diminish. Carbons can be made which will operate satis-
factorily at the higher arc-voltages but such operation is obtained
at the sacrifice of candle-power due to the increased amount of
"sustainers" required at the expense of the quantity of
"illuminants" present. This follows from the fact that the total
amount of flaming materials in the carbons is limited by the
slagging tendency so that an increase of one type of material
must be accompanied by some decrease in the others. Taking all
things into consideration, it will be found that the highest candle-
power, the best steadiness and reliability will be obtained in no
volt multiple lamps when the arc voltage does not exceed 65 and
the current not lower than 6 amperes. For series lamps and
alternating current multiple lamps with auto-transformers, 40 to
45 arc volts gives good service when the current is not less than
10 amperes. In general, however, a flame carbon can be made
to operate satisfactorily commercially over a very wide range of
current and voltage for a given lamp, but at present there is no
one carbon that will meet all conditions satisfactorily in every
way in all lamps.
The current densities within the arc will be found to vary quite
widely. The cross-section which the arc assumes is that which
will maintain the equilibrium between the energy input and the
output of light and heat. The size depends on so many factors,
such as the mutual attraction of the current filaments in the arc
stream, the volume of gas produced by all of the different
chemicals at the various temperatures and pressures, the repulsion
of similarly charged electron, that we can say only that the given
current densities have been found. That they are also satisfactory
is very fortunate. The diffused arc mentioned, is an effect
obtained only occasionally in some direct current arcs ; never in
alternating current arcs. It is of much lower efficiency than the
normal direct current arc which has a positive crater current-
density of the order of 300 amperes per square inch. We have
found the usual tendency to be that arcs of the higher current
densities eive higher efficiencies.
THE FLAME CARBON ARC EAMP 1/9
The use of magnesia blocks in the condensing chamber to pre-
vent globe etching can hardly be considered successful. Toward
the end of the trim-life the blocks become covered with dust
which prevents the complete absorption of the harmful gases and
so permits them to attack the globes. The blocks must be re-
newed from time to time, which introduces an additional cost into
the maintenance account. With properly designed carbons, the
magnesia blocks can be omitted and globe etching does not occur.
Mr. F. A. Vaughn (communicated) : Just one thought
appears in connection with the author's reference in his first
paragraph to the relegation to the scrap pile of old equipment,
especially enclosed arc lamps, and old types of the prismatic glass
reflector. This very point is sometimes a matter of considerable
consequence to illuminating engineers who wish to do their duty
not only to their client in any particular case but to the common-
wealth at large as far as the great movement for conservation of
vision is concerned.
The client naturally wishes to obtain as great an advantage in
the disposal of the old equipment as possible and the suggestion
is almost always presented that these units and these reflectors
are good enough for other installations, and he usually makes
every effort possible on his own part and desires the engineer to"
do so also, to dispose of the equipment to other unsuspecting and
unknowing users of illumination. The illuminating engineer's
duty to his client is, perhaps, to obtain the advantage for his client
in so doing, but his duty to the commonwealth is to use a strong,
heavy sledge hammer on the entire equipment and thus put an end
to its career as a destroyer of human vision, not only as far as
that specific client is concerned but also the entire universe.
It thus devolves upon the illuminating engineer to convert his
client to enough sympathy in the conservation of vision move-
ment to feel willing to allow the whole equipment to be disposed
of for all time.
l80 TRANSACTIONS I. E. S. — PART II
THE ILLUMINATION OF MOTION PICTURE
PROJECTORS.*
J. FRANK MARTIN.
Synopsis: After outlining the operation characteristics and require-
ments of the light sources of projection lamps, the color of light, and the
character of screen best adapted to projecting motion pictures, this paper
is concluded with a brief discussion of the question, Does the motion
picture cause eyestrain? Observation, the author states, indicates that
constant viewing of motion pictures tends not only to develop the semi-
voluntary muscles of the eye but to give them greater endurance and
more rapid action. He adds, though, that pronounced flicker in pictures
may lead to over-stimulation and injury. The discomfort and irritation
sometimes experienced by patrons of motion pictures, caused by unsteadi-
ness of the pictures and defects in the films, indicates so much fatigue
which usually recedes without any resulting injury. The author suggests
an examination of the eyes of lantern operators to determine the nature
and extent of the effect of motion pictures on the eye.
The motion picture is now recognized as an educational factor
second in importance only to the printing press. It is a develop-
ment of the last twelve years and like other great inventions
which possess novelty and meet a demand of the public, the in-
troduction has been accompanied by a disregard of scientific
principles and the necessary standardization to secure the greatest
efficiency.
The magnitude of the industry may be judged by a considera-
tion of the fact that there are more than twenty thousand motion
picture theatres in America, having an average daily attendance
of twenty million people and a maximum demand for electric
current in excess of sixty thousand kilowatts, which about equals
that for lighting a city such as Philadelphia, Boston or St. Louis.
This attendance of these theatres compared with the total num-
ber of people who are habitual readers by means of artificial light
shows, that if the general impression that the motion picture seri-
ously affects the eye is correct, the illuminating engineer is con-
fronted with a problem which is not receiving the attention which
its importance demands.
* A paper read before a meeting of the Pittsburgh section of the Illuminating Engi-
neering Society, April iS, 1913.
martin: illumination of motion picture projectors 181
THE SOURCE OF LIGHT FOR PROJECTION.
From the beginning, wherever electric current was available,
the electric arc has been the only source of light considered. The
requirements for a satisfactory projector iluminant are: first, as
near an approach to a point source of light as is possible ; and,
second, the most intense source of light available. The carbon
arc is the nearest approach to these requirements, and the modern
motion picture projector has practically been built around the
electric arc.
CHARACTERISTICS OF THE PROJECTOR ARC.
Fig. i shows the voltage at the arc on both direct and alternat-
ing current, using a type of carbon which is widely used, at vary-
ing current densities. The arc gap, size of carbons and align-
40
o30
D.C.
/
1
t^S
■-'
FLAN
E A.
c
A.C.
j
5 10 15 20 25 30 35 40 45 50 5S
AMPERES
Fig. i. — Characteristics of projector arcs.
ment of carbons was adjusted for each current value so as to give
the most stable performance. The dotted line in Fig. i repre-
sents a special flaming arc carbon, and the curve is representative
for both alternating and direct current.
ALIGNMENT AND DIMENSIONS OF CARBONS.
Fig. 2 illustrates two methods of aligning the carbons and the
distribution of light resulting therefrom. These settings repre-
sent the extremes between which satisfactory performance of the
arc can be secured. On direct current the top carbon must in-
variably be made the positive electrode of the arc so as to direct
182
TRANSACTIONS I. E. S. — PART II
the maximum flux of light in a horizontal plane. On both direct
and alternating current, the top carbon gives the most light by
virtue of its position in the draught of intensely heated glass
from the arc. The top carbon also burns away faster from the
same cause.
'INCLINED RIGHT AN6LE
Fig. 2.— Alignment of electrodes.
It is very important that the size of the carbon and the relative
diameter of the core and shell be closely regulated to the current
density. Both the stability of the arc and the intensity of light
may be materially increased by varying the dimensions of the
carbons used.
THE MECHANISM OF PROJECTOR LAMPS.
The modern projector lamp retains the elementary construction
and principle of the most primative electric light. Fig. 3 and 4
illustrate the most approved forms in use at the present time. Hand
operated lamps are universally used on account of it being im-
possible to construct a lamp which will automatically center the
rrc in the lens axis and compensate for the wandering of the
arc.
THE PERFORMANCE OF PROJECTOR ARCS.
Fig. 5 shows the relative candle-power of the arc on alternating
and direct currents at varying current densities. Readings were
made in a horizontal plane, the size of carbons, alignment and arc
gap were adjusted in each case to give the steadiest performance.
Attention is called to the termination of the curves for the right
angle and flaming arc at about 27 amperes, beyond which point
the magnetic blow out effect in this method of aligning the elec-
MARTIN : ILLUMINATION OP MOTION PICTURE PROJECTORS 183
trodes has such a value that the arc is lengthened, becomes very
unsteady and consequently unsatisfactory.
For the same current density, all forms of the arc have a much
greater intensity on direct current than on alternating current.
Fig. 3.— Straight line lamp.
Fig. 4. — Angle lamp.
This condition is due to the pronounced crater of the direct cur-
rent arc and the ease with which the flux of light from this crater
may be directed by the alignment of the electrodes.
1 84
TRANSACTIONS I, E. S. — PART II
COLOR OF LIGHT.
The color value of the light used in projection is of small im-
portance unless it is considered from an esthetical viewpoint in
conjunction with photographic processes. However, under pres-
ent conditions the combination of effects arising from the use of
3000
2500
£2000
£ 1500
1000
/
*i
/
§
.*
M
4
¥
4*
?r
0f
W"
<W
u»°J?
&
i
LAM
M0
"
1
2000 4000 6000 8000 10,000
EFFECTIVE CANDLE-POWER
Fig. 5.— Performance of projector arc.
arcs giving a bluish light and very contrasty photography gives
exceedingly harsh impressions. The use of an arc in which yel-
low is the predominating color would make a material improve-
ment.
THE LENS SYSTEM OF PROJECTORS.
Many different combinations of lenses have been experimentally
FILM
08" PROJECTION
[j= .LENS
m
.-Hi
SCREEN-)
INTRINSIC
BRILLIANCY
200 OP'
'-GO OP 5I0C-P-" :470C-P
0.0022 C-P-
0.4 FT. -CANDLES
LOSSES IN PER CENT
70
20
Fig. 6. — The lens system of projectors.
developed but no radical changes have been made in the earliest
form of lens used in the magic lantern. The lens system and the
martin: illumination of motion picture projectors 185
losses therein are illustrated in Fig. 6. It has been built up with
a point sources of light as a basis consequently the low efficiency
of 10 per cent, is not surprising and there is apparently a great
opportunity for improvement.
REGULATION OF THE PROJECTOR ARC AND METHODS OF
CONVERTING ALTERNATING TO DIRECT CURRENT.
Within certain limits of projection which may be described as
the illumination necessary to project an image not exceeding one
hundred square feet in area, direct current through a resistance
is not highly preferable to alternating current supplied through an
auto-transformer, the service voltage being equal in each case at
5000
4000
z3000
z
02000
woo
AjC.-IK)V.&
:.mo"
OfH
ENER
1T0R
5ET-
y
/
/
/
A.CJ
UTOT
<mu
"ORM
*}
/
m
s
~
/ ,
4
■-P
/
s
*s
■
-/
*L.
SPECIAL 15 KW. MOT
GENERATOR SET-
)R-
~^~
'*
2000 1000 6000 8000
EFFECTIVE CANDLE-POWER
10,000
Fig. 7.— Performance of projector arc on alternating to direct curient converters.
1 10 volts. Many forms of self-regulating motor generator sets,
autotransformers and rectifiers, all having a regulation char-
acteristic closely approaching that of a metallic resistance, are in
use. Fig. 7 shows the comparative efficiency of the arc operating
in conjunction with these machines. All the readings were taken
after setting the lamp for the most satisfactory operation.
The majority of motion picture theatres use auto-transformers
of high reactance, which are manufactured and marketed under
various trade names.
For the reason that the use of alternating current gives a very
low efficiency at the arc, requiring heavier current for a given
illumination and an accompanying increase in heat dissipated and
under average conditions will increase the inherent flicker of the
1 86 TRANSACTIONS I. £. S. — PART II
projected image, it is highly desirable to standardize the apparatus
so that direct current is supplied to the projection lamp. A care-
ful analysis indicates that this apparatus should be a motor-gen-
erator set, the motor allowing of substitution for either alter-
nating or direct current at different voltages, and the generator
having a capacity of 3.5 kilowatts, 65 volts, direct current, com-
pound wound and designed to deliver 175 per cent, full load for
a five minute period of each hour. One or more projection
lamps, each connected through a small resistance to give the
necessary ballast, could be supplied from this machine and could
be operated simultaneously for short periods without affecting
the other. The field rheostat and lamp ballast should be placed
conveniently near the operator so that the intensity of light can
be regulated for the varying density of the films. This machine
could be universally used with an improvement in economy and
quality of projection. The relative efficiency of this machine is
also illustrated in Fig. 7.
PROJECTION SCREENS.
The low intensity of illumination, which is less than 0.4 foot-
candles on the average motion picture screen, immediately leads
to the conclusion that improving the illumination is largely a
matter of increasing the intensity of the light source and the
efficiency of the lens system. This conclusion is also apparently
substantiated by the fact that the difference in the efficiencies
of the available reflecting surfaces allows of no great in-
crease in reflecting efficiency. However, due to the fact that
the eye is very much more susceptible to changes in illumina-
tion at low intensities than at those ordinarily experienced, it
is possible to apparently double the screen illumination by in-
creasing the efficiency of the reflecting surface only 10 per cent.
In an effort to satisfy a public demand for more wholesome
and better lighted surroundings, there has been developed a
number of improved screens. There are two distinct types of
these screens : one consists of a smooth surface dressed with
aluminum powder and lacquer; the other is a dressing of the
whitest and most opaque paint obtainable, applied to a smooth
surface or to the back of a large piece of plate glass. The
martin: illumination of motion picture projectors 187
aluminum dressed screen closely approaches the efficiency of a
low grade amalgam hacked mirror and unless the surface is
pebbled or roughened, the image on the screen is not entirely
discernible to a spectator sitting to one side, without the angle
of incidence. This screen must also be mounted in a perfect flat
plane to avoid a serious glare and on account of the metallic tint
gives color distortion. A piece of plate glass, backed with an
opaque white surface, although costly, is the ideal screen. This
construction gives a perfectly smooth surface, fine grained and
gives the highest efficiency allowable in securing satisfactory
definition. White oil cloth stretched on a rigid frame is less
durable but almost as satisfactory.
THE ILLUMINATION OF AUDITORIUMS WHERE PROJECTION
IS USED.
The first consideration is to avoid all sources of light which
will cause glare. Wall brackets fall in this category and should
never be placed in the range of vision, between the spectator and
the screen. Indirect illumination from chandeliers or cove light-
ing is highly preferable as it best avoids glare.
The intensity of illumination should be graduated from the
screen outward toward the rear of the auditorium and should
average about 0.2 foot-candle at the level of the backs of seats.
The value at the rear where the movement of spectators is con-
gested may generally be increased to one foot-candle with satis-
factory results.
POSSIBLE METHODS OF IMPROVING PROJECTION.
Experiments with a view to adapting the flaming arc to pro-
jection have been made and have demonstrated that an increase
of approximately 60 per cent, can be made in the illumination up
to a current density of 20 amperes. The excessive length and
large area of the flame, which give practically all the light, is a
recession from the desired point source and makes it necessary
to use the right angle alignment of electrodes which determines
limitation of the above current value.
Another method allowing of an increase in the efficiency of
the lens system appears to be in the use of a metallic filament
lamps having a short heavy filament arranged with a reflector and
1 88
TRANSACTIONS I. E. S. — PART II
lens as shown in Fig. 8. The lamp may be worked at a consid-
erable over-voltage and, by interrupting the current periodically
in synchronism with the movement of the film, the overshooting
Fig. 8.— Special lighting system for projectors.
phenomena of the lamp will increase the illumination and the
present sector shutter can be eliminated.
DOES THE MOTION PICTURE CAUSE EYESTRAIN.
It is believed that standardization to secure commercial effi-
ciency should be considered secondary to the correction of
physiological defects. The illuminating engineer in his work in
the field of motion picture projection should give particular
attention to the possibility of evil effects accruing to the eyes
of those who witness picture performances.
Before going further, it must be understood that the follow-
ing analysis is not made from the viewpoint of the ophthalmolo-
gist with a full understanding of the principles of physiology
involved, but is rather from the viewpoint of the layman. If
this paper serves the purpose of attracting attention to the sub-
ject, it will have served its mission.
Eyestrain, discomfort and permanent injury may for practical
purposes be divided into two classes : causes which affect the
muscles of the eye, and those which cause deterioration of the
nerve structure.
martin: illumination of motion picture projectors 189
In considering the effect of motion pictures, two sources of
evil are found to be practically absent. First, there is little danger
from intense light entering the eye and there is no glare unless
foreign sources of light are within the range of vision ; therefore
there can be no deterioration of the nerve structure. Second,
when viewing the image on the screen the involuntary muscles
controlling the focus of the eye lens are inactive and there can
be no evil influence on these muscles. This leaves two possible
conditions which may have an injurious effect. The most com-
mon fault is the flicker which involves the basic principle of
motion picture projection. The other fault lies in the common
but unnecessary evil of unsteadiness in the image on the screen,
which is the result of defective methods in the manufacture of
film or in the maintenance of the projector mechanism.
The effect of flicker is confined to the involuntary muscles
which control the action of the iris. The natural action of the
iris is to close very rapidly and open slowly. When looking at a
motion picture the pupil is contracted to a somewhat smaller
opening than when viewing a still picture having the same
illumination. When flicker is pronounced, the pupil may be
observed to tremble slightly as if attempting to follow the fluctua-
tion of light. This may lead to over-stimulation and permanent
injury.
The discomfort and irritation of the eye sometimes experienced
by the patrons of motion picture theatres is due to the un-
steadiness of the picture and blemishes of the film used. These
faults affect the semi-voluntary muscles which control the move-
ment of the eye ball and the discomfort is due to fatigue which
recedes without apparent injury. Observation indicates that con-
stant viewing of motion pictures tends to develop these muscles,
giving them greater endurance and more rapid action. To secure
light on the subject it is suggested that observations at first be
confined to motion picture operators who as a class of trades-
men do not show the evil effects that would be expected.
I90 TRANSACTIONS I. £. S. — PART II
DISCUSSION.
Dr. Eluce M. Alger (communicated) : In discussing Mr.
Martins very interesting paper I must confess at the outset that
my personal experience with moving pictures has been more
theoretical than practical because to me, as to many other people,
they have been so productive of strain and discomfort and head-
ache that I avoid them.
It does not of course necessarily follow that, because they cause
discomfort and fatigue, that they also cause organic damage to
the eyes provided the fatigue is kept within the individuals
physiological limits. That point can only be determined, as Mr.
Martin suggests, by a careful observation of the effects on the eyes
of employees and others whose exposure is a continuous instead
of an occasional one.
Neither do I doubt that even in matters of fatigue and strain
most individuals would with practise develop a great increase in
their ability to compensate for them. But I do feel reasonably
sure that the strain imposed by frequent watching moving
pictures, when superimposed on the fatigue caused most of us
by our customary ocular tasks, very frequently passes beyond
physiological limits and must eventually produce deterioration.
Furthermore it seems to me that many of the factors which
cause fatigue cannot be done away with but are inherent in the
pictures by reason of their being "motion" pictures.
I am inclined to disagree with the author's opinion that there
can be no strain of any account caused by the action of the ciliary
muscles in focussing, since the screen is some distance away from
the eyes and that distance an unchanging one. Even with a
perfectly normal eye the intentness with which one must watch
a moving picture prevents in large measure the complete muscular
relaxation which at frequent intervals should rest our eyes. But
very few of us have normal eyes. Most of us are far sighted or
astigmatic and many of us have muscular defects as well. We
have to focus to see even distant details distinctly and while this
muscular effort need not be very great for the moment the con-
stant rapid changes on the screen make it almost continuous.
In real life we see things momentarily much as a painter would
paint them. Things in the plane for which our eyes are adjusted
ILLUMINATION OF MOTION PICTURE PROJECTORS I9I
are clear and distinct while those out of focus are more or less
blurred, so that we may be only partly conscious of them. In the
moving picture, things, which in nature were in different planes,
are reproduced on the one plane of the screen and so in a sense
are forced upon our consciousness simultaneously. As a result
there are a multitude of distracting details in range of our eyes
at one time and we regard a moving picture with an unwinking
intentness which is very rarely called for in ordinary life.
Neither do the extrinsic muscles of the eyes escape strain.
Even if the picture were an unchanging one the mere matter of
watching it implies the co-ordination of the eyes which depends on
the extrinsic muscles. But in the moving picture motions which
are intended to appear to us smooth and continuous as they were
in life really proceed by a very rapid succession of stops and
starts and the actions are usually completed in a very much shorter
time than they took in real life. Possibly by being less economical
in the matter of film the jerky motion of the figures in the picture
could be largely done away with but in practice we are always
conscious of and instinctively try to follow these movements. As
a result our extrinsic eye muscles are in a state of steady nervous
and muscular tension which is both abnormal and fatiguing.
But the most obvious cause of eye-strain in watching moving
pictures are the rapid jerky motions of the whole picture, the
occasional showers of light flashes, and the variations in light
intensity which give us a sensation of flickering light. The first
two are said to be due to imperfections in the film and can be
done away with by more careful construction, but the flicker
seems to be inherent in the business. I should say that the dis-
agreeable effect of flicker was only partly due to its effect on the
iris. Light falling on the retina causes a stimulation of a centre
in the brain which in turn causes a contraction of the pupil. If
the light remains steady the retina becomes adapted to it and the
pupil dilates somewhat. When the light stimulus is removed the
pupil dilates pretty widely. This whole process requires an
appreciable time and the degree of contraction depends not only
on the brightness of the light to which the eye is exposed but also
on the degree of illumination to which it had previously become
adapted. So that a light of rather low intensity, provided the
4
192 TRANSACTIONS I. E. S. — PART II
room is comparatively dark, may cause marked contraction. But
the flicker of the moving picture applies and removes the stimulus
of bright light much more rapidly than the muscle can respond,
and as a result the pupil remains in a state of rapid oscillation
which is not synchronous with the actual variations in light inten-
sity. The outcome is not only fatigue of the sphincter pupillae
but a state of nerve exhaustion from the too rapid discharge of
nerve impulses.
Another difficulty which has to do with the adaption of the eye
to light is this. All observers are agreed on the retinal fatigue
that comes from having one portion of the retina exposed to light
while the rest of it is in comparative darkness. A strong in-
candescent light in a dark room is too bright to look directly at
while if the room be filled with daylight the bulb seems hardly
visible. In the moving picture one is looking intently at a screen
on which the light may not be intrinsically very intense but, by
comparison with the surroundings, seems very bright indeed. It
would probably be very much easier for the eyes of the observers
if the theatres were lighted in such a way that while no direct light
fell on the eyes the general illumination was as good as was
compatible with visibility of the screen.
Mr. F. A. Vaughn (communicated) : The author's discussion
of eye-strain, discomfort and permanent injury caused by the
motion pictures would, it is thought, lead one to suppose that it
was a matter which could be dismissed without very great con-
cern, as the troubles, when apparent at all, are caused, according
to the author, by undue "unsteadiness of the picture and
blemishes of the film used". While the discussion may be accurate
as far as it covers the case, it is believed that it should be pointed
out that there have been severe cases of ocular disturbances cited
by opthalmologists and the subject is of enough concern to have
received attention in the Calif ornia State Journal of Medicine in
August, 1912, under the subject of "Ocular Disturbances caused
by the Cinematograph" through the authorship of Morton E.
Hart, M. D., San Francisco. Dr. Hart discusses it in part as
follows :
Ocular disturbances due to the cinematograph have, up to the present
time, received practically no mention in medical literature. It seems
strange that this should be the case, for no doubt it has fallen to the
ILLUMINATION OF MOTION PICTURE PROJECTORS I93
lot of almost every oculist, particularly in the large cities to have seen
and treated many patients suffering from this new disease. And there
are very good reasons that there should be ocular disturbances from this
new plaything of the peopie. * * *
The ocular disturbances, classified under the generic term of "cine-
matopthalmia," are really disturbances of vision due to traumatism, and
are matters of degree. The process is the same in all of the conditions.
There are those cases which are merely transient in their disturbance.
When the picture is first thrown on the screen, the individual is incon-
venienced by photophobia and a few tears. He closes his eyes and these
symptoms soon pass away after a few seconds of repose, and the retina
accustoms itself to the new condition of affairs. A further degree is of
longer duration; the retina cannot adopt itself to the fatigue imposed on
it and each time the individual opens his eyes, the symptoms reappear.
It is impossible to continue the spectacle. After leaving the theatre, the
disturbance still persists and in addition to the mild photophobia and
lacrymation there ensues a slight reddening of the conjunctiva. A few
hours, or at least a night's rest, will return the eyes to their normal tone.
In the third degree of disturbance, the symptoms are more severe
and the return to the normal somewhat prolonged. Here the phtophobia,
lacrymation and conjunctivitis persist for several days and in addition,
we have a smarting and itching of the eyes.
In the very severe cases, besides the inflammation of the conjunctiva
with its attendant symptoms of lacrymation and phtophobia, we have
very definite asthenopic symptoms, both accommodative and retinal — the
former due to the ciliary strain and the latter due to a hyperesthesia of
the retina. The distant vision remains normal. Under examination these
patients are found to have no error of refraction or lesion of the fundus.
A case in question may here be cited :
E. R., female, age 16, was brought to me with the following complaint : Eyes burned
and itched and the lids were red, particularly at night. Reading was impossible on ac-
count of blurring of the page. No headaches. This condition would clear up after a
night's rest, to reappear again at frequent intervals.
On examination a slight reddening of the conjunctiva was found and under a myd-
riatic an error of one degree of hyperopia, which was corrected. The near point was nor-
mal, showing no error of accommodation. Of course this was tested before using the
mydriatic. No lesion of the fundus was found. Unfortunately the patient could not be
seen during an attack.
After wearing the glasses for several weeks, the patient reported, stating that the con-
dition had not improved. She was then closely questioned and it was found that it was
her habit to attend a moving picture show at least four times a week after school and
unbeknown to her mother. She was forbidden this amusement and the condition entire-
ly cleared up.
Fortunately these ocular disturbances are not serious and will clear
up under simple collyria and rest.
The question will naturally arise, how can we do away with the cause
of the trouble?
First: The films must be perfect and free from all imperfections.
194 TRANSACTIONS I. Z. S. — PART II
We have all noticed the scratches on the pictures, particularly at the end
of the reels, due to careless handling. When we realize that the average
picture thrown on the screen is about 97,000 times larger than the original
size of the individual film, we can appreciate that even the smallest
blemish on the films will be tremendously magnified on the curtain and
will have a correspondingly bad effect on the eyes.
Second : The illumination must be steady, must not vary and must
neither be too bright nor too dim, for this causes fatigue.
Third : The speed with which the films are turned must be regular.
Any irregularity will have a tendency to cause ocular fatigue.
Fourth : The position of the spectator is very important and should
receive proper regulation at the hands of the authorities. First of all,
there should be no seats placed at the sides of the auditorium. Every
seat should be in direct line with the curtain. This will do away with
the distortion of the picture. Anyone who has had the experience of
sitting on the side, can appreciate the intense strain and fatigue placed
on the eyes.
No seat should be placed nearer than twenty feet from the screen
and further if practicable, depending upon the size of the picture on the
curtain. This will do away with any accommodative effort on the part of
the spectator and thus will reduce the fatigue to a minimum. The nearer
the screen the greater the fatigue, so the seats at the rear of the audi-
torium are the best.
Some people are very susceptible to the influence of motion
pictures, not being able to sit through one performance without
headache or severe eye-strain. It will thus be seen that quite
painful and somewhat serious cases of ocular disturbances have
been caused and probably will be caused in the future by motion
pictures and that, while this may be due to imperfections in their
operation or presentation, these imperfections will probably obtain
in the future as well, and this subject should therefore receive
serious consideration from ophthalmologists, as well as illuminat-
ing engineers, at least until the imperfections are eliminated.
After that it may be determined whether there is still remaining
serious objection to them from the standpoint of ocular comfort.
Dr. P. W. Cobb (communicated) : The motion picture man
has a set of problems pretty much his own. Mr. Martin's paper
touches some of them in a way which interests me. I am sur-
prised at his statement that the illumination on the screen is only
0.4 foot-candle but such surprises are many for one who attempts
to estimate illumination from visual impression. The motion
picture auditorium has to have a low general illumination, just
ILLUMINATION OE MOTION PICTURE PROJECTORS I95
enough for a person whose eyes are light-adapted on his entrance
to the theatre to see to take a place. Mr. Martin places this at an
average of 0.2 foot-candles. This sets the pace, as it were, for
the eye and determines the illumination necessary upon the screen
in order to make the pictures stand out brilliantly. The question
of contrast enters here as an important factor; the contrast be-
tween the picture on the screen on the one hand, and on the other
its surroundings, the seats, walls and ceiling of the theatre. The
surface brightness of these latter is a factor in determining the
visual brightness of the screen, equal in importance to the lumens
per unit area of the screen itself.
Speaking of contrast, I should like to ask if tinting the general
illumination of the room has ever been used to modify the appar-
ent color of the screen. For instance, I should expect that by
tinting the lamps in the auditorium a bluish color, the pictures
could be made to appear yellowish, or by making the lamps
greenish the screen would show a tendency to a rosy pink. A
small amount of tint in the general illumination would be enough,
I should think, to get a pronounced contrast effect on the screen.
It is gratifying to note that the question of eye-strain has been
considered. At 0.4 foot-candles the question of too intense light
cannot merit discussion and glare can be ruled out. What should
be, and is, considered is flicker and unsteadiness of the image.
I cannot agree with what Mr. Martin says as to the effect of
flicker being confined to the muscle of the iris which controls the
pupil.
It has been shown that when the eye is kept in the dark and
illuminated by a momentary flash of light the reaction of the
pupil is delayed for something like half a second. The con-
traction follows and the pupil does not begin to dilate until a
lapse of about 10 seconds. Such movements are altogether too
slow to follow the flicker of a motion picture which has a rate
say of 10 per second or thereabouts. As a matter of fact the
pupil is always undergoing slight fluctuations in size in the
absence of all changes in light, a fact which can be verified by
anyone who cares to examine his own pupil by means of a mirror
and a lens of low power. It would appear to me that the examina-
tion of the pupil under strongly flickering light would be attended
I96 TRANSACTIONS I. E. S. — PART II
with great uncertainties and make it difficult to estimate the
amount of fluctuation and compare it with the normal.
Nevertheless, flicker is undoubtedly disturbing to the eyes, a
fact which was much made use of in getting electric incandescent
lights into use when they were first introduced to the public. The
exact way in which flicker embarrasses the eyes is, I believe, still
not known.
By far the most serious evil of motion pictures seems to me to
be unsteadiness of the image, which I am glad to know is an un-
necessary one. A person leaving a motion picture theatre to go
back to an occupation not especially involving the eyes may, as
Mr. Martin says, find the difficulty which he experienced while
viewing the pictures to disappear rapidly, if he continues to
notice it at all. But it must be far different with one whose call-
ing depends on close application of the eyes. Here the muscles
of the eye, which have been subject to all sorts of surprises in
trying to fix upon a wavering image are taken away from this
only to do more work of a similar kind, reading, sewing or what
not, which calls for the finest sort of co-ordination of the twelve
extremal muscles of the eyes. When one reflects upon the
delicate balance of these muscles, and the incessant and rapid
movements which they must accurately carry out in fine eye-work
it seems impossible that such a shaking up as they get in attempt-
ing to fix an image which jerks about irregularly upon the screen
should not unfit the eyes to a greater or less extent for subsequent
work. It is therefore good to hear that such pictures need not
be, and to feel that they will soon be eliminated from the motion
picture world.
As to the suggestion that observations be made upon motion
picture operators to detect changes in the eyes due to the pictures
I wish to ask does the operator himself watch the pictures at all
attentively? My feeling is that he does not, or at most, watches
them "out of the corner of his eye" only to detect gross defects
in the film or in the action of the machine. It is an open question
in my mind whether he would suffer as much in a week as the
casual visitor to his theatre does in an hour.
The visitor is eager and intensely interested in the story of the
picture (which is a very old story to the operator) and applies his
ILLUMINATION OF MOTION PICTURE; PROJECTORS I97
attention and his eyes closely to the image. It is just this close
attention which whips up the eye-muscles to the task of observing
the image and spurs them on to their own confusion if the latter
be unsteady.
Mr. Edward L. Simon: (communicated) : This paper is in-
deed food for extended investigation and the time is now ripe
for a uniform and scientifically designed and arranged source of
light for projection.
The large film manufacturers have and are expending
hundreds of thousands of dollars for the improvement of the films
and these films when delivered to the exhibitors are mechanically
perfect. This does not apply to the small manufacturers and the
audience is not able to judge between a good and bad film owing
to a defective projecting machine.
The projecting machine of the future must be a presicion
machine mounted on a concrete or some other rigid foundation.
The light should be a cold light and to get rid of the flicker the
screen should never be dark, which is now the case sixteen times
every second. It is not right that a beautiful picture should be
spoiled in its projection by being run through a loose and untrue
projecting machine with a projecting light that is operated
according to the individual operators ideas. It will not be long
before the manufacturers will be compelled to insist upon rigid
enforcements in the way of perfect projecting machines and
proper light source.
The film as used to-day is enlarged one hundred and fifty
diameters. The large productions in the future will not be en-
larged more than fifty diameters and the audience will have the
pleasure of looking at pictures as steady as stereoptican views.
Mr. Louis C. Smith (communicated) : In my experience in
projecting pictures, I have found that considerable attention
should be paid to the regulating of the quantity of light for
different films, slides, etc. I do not apply' this to dense films
entirely, but to various films of varying subjects. Recently, I
projected some cloud scenes, the effect of which when lighted
with all the available energy was so contrasty that it was un-
natural. In general, though, I have found that the general public
like the bright contrasty photographs and motion pictures. More
I98 TRANSACTIONS I. K. S. — PART II
attention seems to be paid to the theme of the picture than the
details of it. This is evidently so partly because defects in the
pictures are so usual.
Although large amounts of money are spent in the moving
picture field, the theatres seldom get good complete projection
outfits ; then again the operator usually does little more than turn
the crank mechanically; he pays little attention, if any, to getting
the best results out of his apparatus.
There are subjects which require great speed and others that
should not be projected so rapidly; usually all are projected
alike. Many things depend directly upon the taste and
experience of the operator.
Some of the most pleasing slides have been projected by a
calcium light. This seems to be due to the yellowness that the
author speaks of. No doubt a yellowish bright light would be
suitable for many films.
TRANSACTIONS
OF THE
Illuminating Engineering Society
Published monthly, except during July, August, and September, by the
ILLUMINATING ENGINEERING SOCIETY
General Offices: 29 West Thirty-Ninth Street. New York
Vol. VIII
MAY. 1913
No. 5
Council Notes.
A meeting of the Council was held in
the general offices of the society, 29
West 39th Street, New York, May 9,
101 3. Those in attendance were: Pres-
ton S. Millar, president ; George S.
Barrows, C. O. Bond, J. W. Cowles,
Joseph D. Israel, general secretary; A.
E. Kennelly, Xorman Macbeth, L. B.
Marks, treasurer; W. Cullen Morris,
W. J. Serrill, R. C. Ware and Arthur
Williams.
The Executive Committee reported
that it had held a meeting April 25.
The business transacted by the com-
mittee is given in the following minutes
of the meeting:
It was decided to accept the invitation of the
Pittsburgh section to hold the 1913 convention
of the society at Pittsburgh in the fall.
After a general discussion of convention
plans President Millar was directed to appoint
a general convention committee. It was under-
stood that the general committee might appoint
such sub-committees as may be required to care
for the various details and arrangements of the
convention.
A draft of the society's conspectus, submitted
by Mr. Macbeth, was discussed informally.
The above report was received and
concurred in by the Council.
Vouchers 1273 to 1314 inclusive,
covering May bills aggregating $1,086.70,
which had been approved by the Finance
Committee were authorized paid. A
monthly report on the society's finances
and membership was received from the
general secretary. The expenses of the
first four months of 1913 were said to
have totaled $2,823.29. The membership
including the additions and defections
presented at the meeting was said to
have totaled 1,363 members. The mem-
bership at the beginning of the year
was 1,335.
Reports on section activities were
received from the following vice-presi-
dents: W. J. Serrill, Philadelphia; J.
\V. Cowles, New England ; Norman
Macbeth, New York; Howard S. Evans,
Pittsburgh, and James Cravath, Chicago.
The Section Development Committee
reported that it had completed a guide
on section management which would
be issued shortly.
The chairman of the Committee on
Papers reported tentatively on the pro-
gram of papers for the 1913 convention.
A brief and informal report on the
progress of plans for the 1913 Conven-
tion was received from President Millar.
The following appointments of the
president to the Convention Executive
Committee were approved : C. A. Little-
field, chairman ; M. C. Rypinski, D. Mc-
Farlan Mo'ore, O. H. Fogg and Thomas
S. Henderson.
Additional appointments to committees
made by the president since the last
Council meeting were approved.
The following committee of five
tellers was appointed to count the
TRANSACTIONS I. K. S. — PART I
ballots of the annual election: G. B.
Nichols, chairman ; L. J. Lewinson, A.
L. Powell, Raymond W. Stafford and
Thomas Scofield.
Mr. L. B. Marks presented an oral
and informal report of the progress of
the plans of the Gas Congress which is
to be held in San Francisco in 1915. Mr.
Marks is the representative of the
society on the committee which is to
plan the Congress.
Mr. L. B. Marks chairman of the
Committee on Factory Lighting Legisla-
tion reported that Governor Sulzer of
the State of New York had signed on
April 17, Bill No. 26 which is entitled
"An act to amend the labor law in rela-
tion to protection of employees operat-
ing machinery, thus creating machinery
and the lighting of factories and work
rooms." A section of this bill relating
particularly to the lighting of factories,
passage-ways and work-rooms was
drafted in accordance with recommenda-
tions made by the committee. This sec-
tion of the bill was printed in the April
issue of the Transactions.
Dr. A. E. Kennelly reported orally
for the Committee on Nomenclature and
Standards. He said that a written re-
port would be forthcoming from the
committee shortly.
Section Notes.
CHICAGO SECTION.
At a meeting of the Chicago Section
in the auditorium of the Western So-
ciety of Engineers, May 14, Mr. M.
Luckiesh of the National Electric Lamp
Association, Cleveland, gave a lecture
on "Light and Art." The lecture, which
was accompanied by a series of well plan-
ned demonstrations showing the effect
of direction, and quantity and quality of
light on objects of art and design,
proved to be intensely interesting to the
forty members and guests in attendance.
The lecture had been given before meet-
ings of several other sections of the
society.
At the June meeting, which will prob-
ably be held on the 18th, Mr. Arthur
J. Sweet will present a paper on car
lighting.
NEW ENGLAND SECTION.
The May meeting of the New Eng-
land Section was postponed. No date
or program has been scheduled for a
June meeting.
NEW YORK SECTION.
Before a largely attended meeting of
the New York Section, May 8, Mr.
Bassett Jones, Jr., gave a very interest-
ing lecture and series of demonstrations
on theater lighting. The meeting was
held in the Clymer Street Theater,
Brooklyn, through the courtesy of the
proprietors. The uusal monthly dinner
at Keene's Chop House, preceded the
meeting.
PHILADELPHIA SECTION.
The May meeting of the Philadelphia
Section, in the Franklin Institute, May
16, was addressed by Dr. Herbert E.
Ives on "Some Home Experiments in
Illumination from Large Light Sources,"
and Mr. C. A. Peterson on "The Design
of Combination Fixtures." There was
also an exhibition of the latest types of
residential glassware.
The Section will have a June outing
which will be a joint outing with the
Philadelphia Section of the American
Institute of Electrical Engineers on the
grounds of the Philadelphia Electric
Company Athletic Association, on
TRANSACTIONS I. E. S. — PART I
Saturday afternoon, June "th. There
will be a baseball game between mem-
bers of the Illuminating Engineering
Society and the American Institute of
Electrical Engineers. Prof. Franklin of
Lehigh University will give a lecture
on "Baseball Curves." Supper will be
served on the grounds
On June 20 the section will hold a
short business meeting and dinner at
the Engineers' Club instead of a regular
meeting. There will be some short
speeches but no paper.
PITTSBURGH SECTION.
A meeting of the Pittsburgh Section
was held May 16. Mr. J. L. Minick of
the Pennsylvania Railroad Company read
a paper on "Passenger Car Lighting."
The paper appears in this issue of the
Transactions.
For the June meeting Messrs. Ward
Harrison and E. J. Edwards of the
National Electric Lamp Association are
preparing a paper on the lighting of the
new office buildings of the National
Electric Lamp Division of the General
Electric Company, at Nela Park, Cleve-
land.
New Members.
The following applicants were elected
members of the society at a meeting of
the Council, May 9, 1913 :
Amkine, T. H.
General Electric Company, Harri-
son. N. J.
Bergman, Rudolph L.
Salesman, Benjamin Moore & Co.,
231 Front Street, Brooklyn, N. Y.
Bern hard., Albert H.
1070 Bedford Avenue, Brooklyn,
N. Y.
Bryant, John Myron
Asst. Professor of Electrical En-
gineering, University of Illinois,
Urbana, 111.
Davidson, John M.
Civil and Sanitary Engineer, Ameri-
can Sheet and Tin Plate Company,
1224 Frick Building, Pittsburgh, Pa.
Ehrlich, Howard
Associate Editor, Electrical Re-
view and Western Electrician, 608
S. Dearborn Street, Chicago, 111.
Ellis, Edgar J.
President United Electric Const.
Company, 1727 Sansom Street,
Philadelphia, Pa.
Frith, Andrew M.
Burnham-Frith Electric Company,
MacDougall Avenue, Edmonton,
Alta., Can.
Froelich, J. M.
Arc Lighting Engineer, Duquesne
Light Company, 435 Sixth Avenue,
Pittsburgh, Pa.
Gray, A. A.
Managing Editor, Electrical Re-
z'iezu and Western Electrician, 608
S. Dearborn Street, Chicago. 111.
Grondahl, L. O.
Instructor, Carnegie Institute of
Technology, Pittsburgh, Pa.
Henderson, R. G.
District Manager, General Electric
Company, 30 Church Street, New
York, N. Y.
Ives, Arthur S.
Ives & Davidson, 84 William Street,
New York, N. Y.
Kaulke, Johannes
Electrical Engineer, General Elec-
tric Company, Sussex and Fourth
Streets, Harrison, N. J.
Kelley, J. B.
Salesman, Frank H. Stewart Elec-
tric Company, 35 N. 7th Street,
Philadelphia, Pa.
TRANSACTIONS I. E. S. — PART I
Kelly, Clarence B.
Chief Estimator;, United Electric
Const. Co., 1708 Sansom Street,
Philadelphia, Pa.
Latta, J. E.
Associate Editor, Electrical Re-
view, 608 S. Dearborn Street,
Chicago, 111.
Landerdale, Jesse E.
Sales Engineer, National X-Ray
Reflector Company, 235 West
Jackson Boulevard, Chicago, 111.
Mason, Frank L.
Instructor, Dept. Electrical En-
gineering, Columbia University,
New York, N. Y.
Mauser, R. H.
Engineer's Assistant, Consolidated
Gas Company, 124 East 15th Street,
New York, N. Y.
McNeil, R. S.
General Electric Company, Harri-
son, N. J.
Otto, William G.
General Sales Manager, Walker
Electric Company, 2338 Nobb Street,
Philadelphia, Pa.
Rowland, Ernest W.
Chief Bond Inspector, Public Ser-
vice Railway, Newark, N. J.
Rypinski, M. C.
Manager Detail and Supply Dept,
Westinghouse Electric and Mfg.
Company, 163 Broadway, New
York, N. Y.
Truitt, Thomas Gibe
Imperial Electric Company, 1022
Arch Street, Philadelphia. Pa.
Watkins, Howard E.
Vice-President and Designer, The
Enos & Watkins Co., 36 West 37th
Street, New York, N. Y.
Sustaining Member.
The United Electric Light and Power
Company, New York, was elected a sus-
taining member of the society at a
meeting of the Council, Ma}' 9, 1913.
Joint Meeting I. E. S. and A. G. I.
The following announcement has
been issued by a special committee of
the society, which is arranging for a
joint meeting with the American Gas
Institute :
"A joint session of the American Gas
Institute and the Illuminating Engineer-
ing Society, will be held during the third
week in October at the annual meeting
of the Institute in Richmond, Ya.
While it is true that a number of gas
men are active and influential in the
work of our society, still a regret is fre-
quently expressed that gas men and gas
companies generally, do not take a
more active interest. The gas business
is so intimately bound up with our pur-
poses and objects that it seems un-
necessary to dwell on the value of our
society to that great industry.
Your committee believes that this
joint meeting will furnish an excellent
opportunity for the arousing and
stimulation of such an interest. * It
urges on the membership of our society
the importance of a large attendance at
this joint meeting, and of an abundant
discussion of the papers presented to it.
We are notified by the Committee on
Papers that the following papers have
been secured :
"Some phases of the Illumination of
Interiors," by Preston S. Millar, Secre-
tary, Electrical Testing Laboratories,
New York, N. Y. A lecture-demonstra-
tion employing minature rooms to
illustrate several well known types of
TRANSACTIONS I. E. S.— PART I
lighting installations, indicating their
peculiarities and good and bad features.
"The Importance of Direction, Qual-
ity and Quantitative Distribution of
Light in Illumination," by M. Luckiesh,
assistant physicist, National Electric
Lamp Association, Cleveland, Ohio, a
lecture-demonstration chiefly by sub-
jects taken from the fine arts.
"Gas Lighting of Interiors," by C. A.
Luther, Illuminating Engineer, Peoples
Gas Light and Coke Company, Chicago.
111., a paper dealing with the lighting of
interiors by gas, and illustrating the
manner in which gas is used to obtain
the results explained in Mr. Millar's
demonstration.
"Street Lighting by Gas,'' by F.
V. Westermaier, Engineer, Welsbach
Street Lighting Company of America,
Philadelphia, Pa., a paper dealing with
street lighting from the standpoint of
the most modern methods.
This will evidently be an entertaining
and instructive session. The date of
the joint meeting will be announced
later.
1913 I. E. S. Convention.
The seventh annual convention of
the Illuminating Engineering Society
will be held in Pittsburgh, September
22 to 26 inclusive. A committee of
which Mr. C. A. Littlefield, 55 Duane
Street, Xew York, is chairman, is en-
gaged in formulating plans for the
biggest and most successful convention
yet held by the society.
TRANSACTIONS
OF THE
Illuminating
Engineering Society
MAY, 1913
PART II
ck
<K
Papers, Discussions and Reports
[ MAY, 1913 ]
CONTENTS - PART II
Street Lighting of Greater New York. By C. F. La-
combe 199
Illumination of Passenger Cars. By J. L. Minick 214
H
STREET LIGHTING OF GREATER NEW YORK.*
BY C. F. LACOMBE.
Synopsis: Included in the following paper is a description of the plan
or system of street lighting of the City of New York. With its accom-
panying illustrations the paper presents a comprehensive outline of the
progress which has been made, particularly within the past decade.
In appearing before you this evening in an endeavor to de-
scribe the street lighting of the City of New York and the efforts
made to improve it, and to make it attractive and artistic, my
attitude is not that of a Philistine, satisfied fully with the engi-
neering and efficiency results attained. Nor is it my desire to
impress that idea, for the results so far reached are not entirely
satisfactory. More than that, however, the endeavor has been
made within imposed limitations to make the lighting agreeable
and artistic, and on this point it is hoped your interest and
sympathies may be enlisted by the description of the work that
is being done.
In order to make the situation clear, it is necessary to briefly
describe the limitations under which one has to work in the
city government. Like all artistic or utilitarian things, illumina-
tion is. a matter of money to a certain degree. Since 1905, the
legislature has limited, properly, the prices to be paid for illumi-
nating units, and the Board of Estimate and Apportionment, in
the budget, has limited from year to year the number of units of
illumination that may be added by setting the amounts of money
that can be used in each borough. There are further limitations
in connection with opening streets, etc., which restrict the time
of construction, particularly in Manhattan and Brooklyn, so that
only a certain amount of physical work can be done each year.
In consequence, progress in improving lighting is slow and con-
servative to avoid frequent changes, and should be so, particularly
in the older sections of the city. None of you who know the
illuminating art is willing to say the last improvement has been
* An address delivered t efore a joint meeting of the Municipal Art Society and the
New York section of the Illuminating Engineering Society, February 21, 1913.
200 TRANSACTIONS I. E. S. — PART II
made, or that the last unit of illmuination has arrived. You also
know how necessary it is not to jump at conclusions, or adopt
new units without careful study and trial, and in this term,
units, is included both the light source, its reflectors and diffusing
media. From this you can see that we are limited to the
use of standard tried lighting units and that new ones cannot be
adopted without a working trial, for the reason that the illumi-
nating service purchased by the city must not be lessened or
impaired.
The city contracts with public service corporations to supply
it with certain illumination and illumination service on its streets.
That means that the illumination is to be given by certain proved
lighting appliances, or, in other words, proved units, and that
these appliances or units are to be kept in a condition giving as
nearly as possible perfect operation. The term "perfect opera-
tion" is the definition of "good service." It depends on many
elements and great attention to detail. In 1888, when first en-
gaged in electrical work, a small handbook was given me which
said "The best electric system is the man who runs it" — a crude
way of saying the best illuminating system depends on the
efforts of the men who run it. This was in the days when there
were many different so-called systems of producing electricity
for illumination work. Now, the efforts of the men who operate
a lighting system largely determine whether the householder or
the community get good or bad service, and the clumsy epigram
quoted is just as true now as then.
Good service in electric and largely in gas lighting, depends on
the active strength and reserve capacity of the central station and.
its relays, the perfection and reliability of the distribution system,
the detailed care and attention given to the most reliable lighting
appliances obtainable for the system, and last, but not least, the
discipline, co-ordination and co-operation enforced between the
above elements as established and maintained by the man in
charge. This man and his loyal assistants may really be called the
slaves of the lamp, and those of us who have been such slaves,
know the strange fascination of this work and its demand for the
sacrifice of all else to it. But that fascination and devotion mean
good service, and successful lighting service can never be obtained
V
^
a b c
Pig. I.— Gas lamp posts: (a) Old gas iron type in use about fifty years. Designed for open
gas flame o/ about 13 candle-power. Weigbt of lantern about 14 pounds, (b) Early
mantle type, using lantern weighing 32 pounds. Average candle-power 35 to 40.
1 c) Present mantle type. Center of illumination is 9' 6" above the ground. Average
candle-power 35 to 40.
ivWEf
Fig- 2.— Arc lamp posts: (a) Bishop's crook type. Design embodies suggestions of Stanford
White. Lamps trimmed from post or by tower wagon, there being no lowering device.
(b) Lyre top post, as used in center-parked plots on broad streets, such as upper Seventh avenue.
(c) Mast arm post, used to avoid tree interference. Used at curb to overhang the street, in
some cases in connection with the lyre top, as in upper Seventh avenue and upper Broadway.
lacombe: street lighting OF NEW YORK 201
without it; nor can steady, continuous illumination— without
which any attempt towards agreeable or artistic lighting is futile—
be obtained.
Assuming this to be obtained, then, let it be asked what choice
is offered in the way of lighting appliances for service. It is
apparent on a moment's thought, that such appliances must be
well tried and standard. New appliances cannot be used on regu-
lar service. They must be given long service tests in by-ways and
unimportant sections, often improved and re-tried until they be-
come reliable and of real value. Of such available appliances
to-day, there are the arc lamp of standard design for both
classes of current supply, and for both multiple and series
systems ; the incandescent lamp, in various intensities, with
carbon, metallized and metal filaments, for similar current supply
and distribution systems ; the mantle gas lamp with either vertical
or inverted mantles, increasing the number of mantles for in-
creasing illumination; and, naphtha-vapor mantle lamps, giving
the same illuminating effect as vertical mantle gas lamps. In
addition, but not available in all parts of the city, are the high
candle-power arc lamps using a metallic electrode, with which
much can be done. In Manhattan, where the greatest congestion
exists, long burning white flaming arc lamps, which afford still
higher illumination for general street lighting are now being tried.
A limited number of these lamps are being installed on Broad-
way, above 47th street and on 7th avenue, above 34th street. This
list of lamps, then, comprises the lighting units which must be
used at present.
This city differs from others in that low and high tension
electrical distribution are both used on an extensive scale. In
Manhattan the lighting is all on low tension multiple circuits,
both direct and alternating. In by far the larger part of Brook-
lyn and all the other boroughs the street lighting, with a few
minor exceptions, is by means of high tension direct or alter-
nating current series circuits.
Gas supply for mantle lighting is available throughout the
greater city, except in a few remote sections. Naphtha-mantle
lamps are still used for frontier lighting, where neither gas nor
electricity is available. Gas lighting was formerly used in parks.
202 TRANSACTIONS I. E. S. — PART II
but is now being abandoned for obvious reasons, in favor of the
tungsten incandescent lamps. Of these various lighting units, in
the greater city, there are: 19,180 standard enclosed arc
lamps, 17,991 incandescent lamps, 78 flaming lamps, perma-
nent and on trial ; 44,653 single mantle gas lamps, 28 inverted
mantle gas lamps on trial, 1,816 naphtha-vapor lamps. In all there
are 83,746 lamps furnished by 28 lighting companies. Expressed
empirically, the lamps give a horizontal illumination approxi-
mately equal to fifteen million candles.
For the purpose of street illumination, these lamps are used
in the following manner :
Enclosed arc lamps, reinforced by flaming arc lamps, are in-
stalled at points of great congestion, as at Times Square, Man-
hattan.
Arc lamps are used on main avenues and business streets,
practically over their whole length in Manhattan, and as far as
necessary in other boroughs, and then merged into tungsten in-
candescent electric or gas mantle lamps.
Gas lamps are generally used in the residence districts and on
unimportant streets, little used at night, although of late, for
many reasons, the tungsten lamp has made great inroads on this
territory.
Ten years ago last month, when the engineering charge of the
illumination of Manhattan and the Bronx was put in my hands, it
was found to be inadequate, unsymmetrical and out of
date; the arc lamps on the main avenues were insufficient
and often spaced irregularly, too far apart and not in
symmetry in line or in height, except where standard fixtures
were employed. Open flame gas lamps were in the majority,
with gas mantles and arc lamps often mixed in with them. In
other words, there was little, if any, system of lighting.
As soon as possible, this was corrected. First, prices were
materially reduced by continuous agitation, and second, a plan of
lighting developed. All open flame gas lamps were discontinued.
Arc lamps were practically doubled in number at street and ave-
nue intersections and symmetrically arranged as often as
possible on street house lines; so that the illumination was made
as uniform as it could be with the funds allowed. Mantle ?as
(K
<y>
Fig- 3- — (a) Bracket type ; for use on buildings in narrow streets downtown, where a post would
take up material width of limited sidewalk, (b) Twin lamp posts, as used in Fifth avenue,
(c) Reverse scroll bracket post, as used on lower Seventh avenue and elsewhere ; also used with
flame arc lamps.
Fig. 4.— Arc lamp posts: fa) High type ; used with powerful flame arc lamps in squares and open
flames, as for example Times Square, t,ong Acre Square, etc.; height usually 45 feet, with globe
of lamp about 39 feet from ground, (c) Ornamental pole, designed by McKim, Mead and White,
for flame arc lamps used around new Municipal Building, Center and Chambers streets.
LACOMBE: STREET LIGHTING OE NEW YORK 20.3
lamps were put on existing lamp posts in all residence streets ex-
cepting a few quarters where arc lamps were already installed.
This arrangement as a whole has proved efficient and has been
followed with improvements as they were worked out. The
result may be seen in the enlarged system of uniform distribution
of light sources in long straight parallel lines, exemplified best,
first by Seventh avenue above Central Park and Fifth avenue
from Washington Square to 6oth street, and now by almost all
the avenues in Manhattan, from west to east, running north to
south, as well as on the main streets running east and west.
Late in 1905, the engineering charge of the illumination of the
greater city was put in my care, as chief engineer, and early in
1906 such control was formally assumed. The same general
system as used in Manhattan and the Bronx was put into effect
in the other boroughs and the bureau organization in these bor-
oughs made to conform to the simple system of develop-
ment in illumination just described in Manhattan. It was
modified in some ways, for in the outlying sections of
Brooklyn and in Queens and Richmond the problem was
quite different. In these boroughs it was the illumination
of small towns and centres, with long connecting roads
through truck farming districts from centre to centre. These dis-
tricts are lighted by means of far flung alternating and direct cur-
rent series systems of electrical distribution overhead on poles,
with all the limitations of such suburban systems and companies.
The same rule was worked out in symmetrical, continuous and
uniform illumination with arc and tungsten lamps, and there are
now established certain lines of well lighted roads from the City
Hall in Manhattan, north to Yonkers, east to Nassau County, and
central Long Island, as well as its north shore, southeast to
Rockaway, south to Coney Island, and excepting the ferry, to the
southernmost point of Staten Island. Cross and inter-connecting
roads through country districts are also carefully illuminated.
Other projects on this line are now building or planned, such as
the Boston Road and Pelham Parkway and Park to Pelham and
the south shore of Connecticut, another route via the Eastern
Parkway, Brooklyn, to the Merrick Road and the south shore of
Loner Island.
204 TRANSACTIONS I. E. S. — PART II
This general system is now established on such basic lines that
it can be developed consistently and without duplication of equip-
ment expense using the present lighting units. Until congestion
spreads from Manhattan, the lower portion of the Bronx and
Brooklyn, to the outlying sections, to the extent that a consider-
able change from overhead line construction to underground sub-
ways becomes necessary, the development need only be to higher
units and not a change in systems or equipment to any consider-
able extent.
A certain handicap will exist for some years in obtaining the
best distribution of illumination in the suburbs on account of the
overhead lighting lines and lighting being kept on one side of the
street to avoid duplication of pole lines ; but this, in time, will be
removed.
In the congested area of Manhattan and in the Bronx, where
underground distribution exists, the locations of the lamps on the
main avenues are secured, and to realize the full perfection of the
present plan, only higher sources of illumination are neded, and
it is to that end the experiments with flaming arc lamps on Broad-
way and Seventh avenue are being made.
In this development care has been taken to avoid freak, or too
accentuated lighting, on particular streets, as has been the case in
some cities. The attempt has always been to keep to uniform,
agreeable lighting, avoiding light sources of high intrinsic
brilliancy and consequent glare. Where high intensity sources are
used they are carefully diffused or kept out of the range of vision.
It is all a slow, but so far, sure progress of evolution, develop-
ment and education. The results of the original plan of ten
years ago now show in Manhattan, the Bronx and Richmond, and
are beginning to show in the thoroughfares of Brooklyn and
Queens. The mixture of lighting units has been generally re-
moved or is in process of removal. In the last ten years, the
illumination of the greater city, as specifically defined, has in-
creased, under this system from seven million candles to fifteen
million candles, or an increase of 118 per cent.
It is a peculiar and somewhat odd fact about all this develop-
ment, that except in a few isolated instances, no one in this city
seems to realize the improvement made in its lighting in later
y>
h
^
* !
Fijj. 5 —Tungsten lamp posts: (a) First ornamental tvpe. French design: used in Central
Park and on Riverside Drive, (b) Present ornamental type, designed by Henry Bacon
for the Municipal Art Commission ; used in park roadways, (c) Ornamental type with
diffusing globe, designed bv Bacon, used in special places in parks, for example. The
Mall. Central Park, (d) City ornamental type, now used with 500-watt tungsten lamp,
as shown, at Sherman Square, 59th street and Fifth avenue. See also Fig. 12.
Fig. 6. — (a) Municipal Art Society post ; designed by Mr. Ciani and presented to the city
by the Municipal Art Society, (b) Astor post ; designed by Henrik Wall in and pre-
sented to the city by the Astor estate. Two located in front of the Astor Hotel, Broad-
way, (c) D. A. R. post; designed by Allen G. Newman and presented to the city by the
Daughters of the American Revolution ; to be placed at 72nd street and Riverside Drive.
(d) Pennsylvania Station post; installed around the Pennsylvania Station and adjacent
postoffice.
LACOMBE : STREET LIGHTING OF NEW YORK 205
years. Few newspapers have ever mentioned it, and then but
casually; no realization that a logical engineering scheme of
development is taking place has been noticed, although it has
been inherently correct enough to have been supported by the
commissioners of four political administrations of the city gov-
ernment. It is only fair to say that the technical journals and
societies have discussed it somewhat, and a number of other
cities have noticed and studied the lighting and fixtures in use
here with a view to their use in their own cities. In some cases
our system has been adopted by them.
Even in this, a sort of peculiar pride is taken ; for this city,
FIFTH AVENUE
Present Installation, Two 450-Watt Arc Lamps per Post
I 1 Over 0.3 Fool-candle V77\ °'3 " "•' F<">,-«ntlle ; ^%j 0.1 • 0.03 Fool-candle
I 0 03 - 0 01 Foot-candle
Fig. 7. — Diagram showing lamp locations and isointensity curves; two 450-watt enclosed arc
lamps per pole; Fifth avenue; length of block 264 feet.
rushing on in its career to the goal of the greatest city of the
world, has little time to do much but growl at real and fancied
obstructions to its progress. In consequence, to keep up to the
extension of the lighting demanded, the lighting bureau of the
department has had little time to do anything but work, its efforts
being devoted to making up for the lack of development in the
past, and hurrying forward to meet the present demands of a
greatly congested world centre. This is really the first address I
have been able to make on this subject.
In a negative way, the work has been noticed ; for example.
206
TRANSACTIONS I. t. S. — PART II
since the avenues have been doubled and tripled in illumination,
we have heard a complaint that the side streets in Manhattan and
the Bronx are dark, and such complaints are always listened to
attentively, because they help the engineering scheme by impress-
ing the political and financial powers that be. The side streets
are comparatively dark, but in 1904, when the mantle gas lamps
replaced the open flame gas lamps, these streets received more
than three times the old illumination, so one can imagine what
that illumination was. It may be truthfully said that were this
city thrown back in a day to the lighting of ten years ago, the
contrast would be so extreme there would be a riot, almost, on
450-Watt Arc Lamps
I 0.3 • 0.1 Foot-candle V/ZZ\ "■! ■ 0-°3 Foot-candle Rx>£3 0.03 - 0.01 Foot-candle
[ ] Over 03 Foot-candle
ig. 8. — Diagram showing lamp locations and isointensity curves; mast arm and lyre top
posts; 450-watt enclosed arc lamps; upper Seventh avenue; length of block 264 feet.
account of lack of illumination. In the elapsed time the increase
has been slow, and, in consequence, not perceptible, except from
year to year.
It is desired to emphasize the fact that under the engineering
scheme all new lighting is carefully worked out mathematically
and geometrically, on lines of proper illumination design. A
temporary equipment is then erected, movable, if desired, and the
final effect obtained by the actual trial and observation on the
street. All suitable kinds of reflecting and diffusing devices and
glassware are tried until the best result is reached. This example
is then measured and plotted photometrically and checked, or
corrected, as the case may be. Lighting is no longer installed
empirically or by guess work.
r
fth avenue twin arc lamp posts (see Fig. 3): location shown in Fig. 7; night
photograph.
Fig. 10.— Upper Seventh avenue ; mast arm and lyre top posts (see Fig. 2); location as
shown in Fig. 8 ; night photograph.
ii. — Broadway lighting with reverse scroll posts shown in Fig
in Fig. 13; night photograph.
locations shown
Fig. 12. — Plaza and Sherman monument, 59th street and Fifth avenue, illuminated by flaming
arc lamps, enclosed arc and tungsten lamps, night photograph; tungsten lamps in the fore-
ground and flaming arc lamps to the left beyond the range of the camera.
lacombe: street lighting OF NEW YORK
207
The newer lamp posts used for the improved lighting are also
designed with great care. Even before the Municipal Art Com-
mission took up this matter in a systematic and effective manner,
as they did some years ago, the New York Edison Company had
anticipated the demand for more attractive fixtures and has been,
and still is pre-eminent in this regard.
It is well to state here, as one of the working conditions, that
with few exceptions the arc lighting posts and fixtures belong to
the companies throughout the greater city, the gas lamp posts all
belong to the city, as do the tungsten lamp posts in the streets and
BROADWAY
Enclosed Flame Arc Lamps
[ — ~\ Over 0.3 Foot-candle \7J\ 0-3 ■ 0.1 Foot-candle ' | 0.1 - 0.03 Foot-candle
Fig. 13.— Diagram showing lamp locations and isointensity curves 450- watt
enclosed flame arc lamps; Broadway.
parks on underground service. The lanterns and lamps them-
selves belong to the various companies. The posts now supplied
by the lighting company are shown in the accompanying illustra-
tions. They were first drawn, then life size plaster models made
and revised, then the patterns developed and corrected until a
satisfactory and harmonious result combining artistic effect with
engineering construction was attained. These posts were then
submitted to the Art Commission in every case and their
criticisms embodied in the final result.
208 TRANSACTIONS I. E. S. — PART II
The city's posts also go through this process. In one case the
Art Commission had designs drawn and paid for of both the
lantern and posts for tungsten lamps for Central Park. This
design has been used extensively throughout the city since and
has been copied by other cities. The new posts for side street
lighting will also be submitted to them, when funds are received,
and it is hoped that the old ugly gas lamp post will shortly dis-
appear from Manhattan.
It is natural that the attempt to make the lighting and fixtures
artistic should begin, like the improved lighting system, in the
borough of greatest congestion, where it is needed most, and
where the lighting is required to be not only useful but strong,
uniform and agreeable.
Manhattan is fortunate, in the engineering sense, in being sup-
plied with its lighting service by a low tension multiple system,
which is flexible and very adaptable to artistic effects with safety.
The lighting current is distributed by underground lines, and
with energy supplied from the greatest electric generating station,
gives the best obtainable service.
It has been my attempt so far to give you, as briefly as possible,
the salient points of the conditions under which lighting work is
done in this city. The endeavor has been made to be non-
technical. It is necessary to remind you that the illustrations
showing night lighting do not at all show the effect of the same
lighting on the eye, so far as the source is concerned, so if you
will compare the surfaces illuminated and not the source, you will
have a more accurate comparison.
It is my pleasant duty to acknowledge how much I am indebted
in accomplishing this work so far, first, to the support of the
commissioners of the department; second, to the collaboration
and assistance of the engineering and service side of The New
York Edison Company, led by Vice-president Lieb, the dean of
all central station men, and to Mr. Rhodes, in charge of the arc
lighting department, who has more than done his half of the work
with me in this development; third, acknowledgment is also due
to my own men, particularly to the general inspector of street
lighting, whose accuracy and great attention to detail have done
much towards the genera result.
Fig. 14. — Central Park Mall ; ornamental ball tungsten lamp posts (see Fig. 5) niglit
photograph ; 60 candle-power lamps.
Fig. 15 — Central Park road lighting; tungsten lamp posts; (see Fig. 5-b) night photograph;
60 candle-power lamps.
Fig. 16. — Side street illumination; experimental tungsten lamp posts; locations
shown in Fig. iS; night photograph.
Fig. 17. — Public library; lighting from enclosed arc lamps on the opposite side of Fifth
avenue; approximately 0.1 foot-candle on face of building; night photograph.
LACOMBE: STREET LIGHTING OF NEW YORK
209
In addition to the progress made, it has also been attempted to
show in this paper, in a limited way, some of the diversified
problems of the engineer in the lighting of this great city. He
must proportion the lighting to the needs of the various streets
or sections, and to their importance, due to greater or less use.
He must lay out a system capable of great increase without ex-
pensive change of equipment, or the contracting companies will
object. He must lay it out on economical and efficient lines and
obtain judicially fair prices within limits, or the city administra-
tion will object. He must try to eliminate glaring lighting, or
the Illuminating Engineering Society will protest and he must
make it agreeable and as artistic as possible or the Municipal Art
Societv and Commission will criticise.
113TH STREET
150-Watt Tungsten Lamps, 113 Feet Spacing
^J 0»ec 0.3 Foot-candle f\^ 0.3-0.1 Foot-candle |\^| 0.1-0.03 Foot-candle
0.03-0 01 Foot-candle
Fig. iS. — Ump locations and isoititensity curves new tungsten lamp posts
for cross streets.
Agreeable and successful lighting is a combination of the efforts
of all the different interests mentioned. It is not only a matter
of artistic posts, but also artistic lighting, if I may use that term.
To produce the best results, one must have the support of the city
and the contracting company. The posts and the lighting must
both be artistic and agreeable. The position, height and design
of the supporting post have much to do with the effect of the dis-
tribution of the lighting, as well as its cost. It is in these latter
details the artist and engineer can well work together. Within
limitations, the artist must not demand too extremely, artistic a
design in lighting; it is too costly except in a few isolated in-
210 TRANSACTIONS I. E. S. — PART II
stances. In the general lighting of a great city, the useful side
must have as great weight as the artistic.
The engineer also should not cling too closely to the most
efficient and economical devices which obtain only the greatest
illumination, at the lowest cost, for the lighting must be made
agreeable to the eye. In street lighting, unless surrounding build-
ings are to be specially illuminated, the useful rays are those that
can be directed towards the ground. To make this agreeable, the
point source of an arc lamp, for instance, must be made into a
ball of softened light by diffusing globes or shades, and this, so
far as is possible, thrown towards the ground by either interior
or exterior reflectors. This practise reduces efficiency to a cer-
tain extent by the absorption of the light, and demands either
closer spacing or more powerful sources of illumination at
greater heights from the ground ; it is, consequently, more expen-
sive. So far, it has been used mainly in Manhattan and there,
principally on its main avenues.
In other locations, one may see the naked source of the light
which, in turn, means glare that is not as agreeable, although
very useful and less expensive. Considerable success in suburban
sections of the city has also been attained by abandoning the in-
tense arc unit and using three to four 60-70 candle-power
tungsten lamps, and so obtaining better distribution, with less
glare, and at about the same cost.
There is nothing new in this plan of lighting. The ancient
torch bearer always held his torch as high as possible to keep the
glare out of one's eyes and to throw the rays over as great an
area as possible. The engineer uses his new facilities for lighting
in the same way, puts them as high as possible, and throws them
over the greatest area he can. He only adds to the torch bearer's
efforts, the diffusing globes and the downward reflectors.
Artists, too, can take advantage of the many possibilities of the
new lighting devices, both in general and decorative illumination.
Beautiful color schemes are possible with small colored lamps set
in mosaic design, both massed and in outlines, on walls of build-
ings. As a substitute for advertising with glaring yellow flame
lamps, the use of mosaic lighting would be vastly better and just
as effective. Even general illumination, as given by the high
lacombe: street lighting OF NEW YORK 211
powered flame lamps can be tinted agreeably in contradistinction
to dead, cold, white light. In other words, artistic lighting is as
possible as any other form of art.
This address would lose its point if it failed in suggesting that
the artist and engineer meet each other half way, and, by com-
bination, produce a joint result, obtaining good, economical light-
ing, consistent with artistic standards. In this way, lighting can
be designed which will command complete approval from all
the points of view that may be invoked to judge it. If such com-
bination is obtained, then I can say that five years of co-operation
will make Manhattan Island first, and the greater city next, the
best and most beautifully lighted city in the world.
DISCUSSION.
Mr. Clarence E. Clewell (communicated) : The author
has pointed out in a most interesting way that the progress in
the lighting of New York has been made by keeping two main
points in view, namely, first to avoid accentuated lighting in par-
ticular streets as has been the case in some cities on the one
hand, and second, to secure uniform agreeable lighting on the
other hand, thus avoiding light sources of high intrinsic bril-
liancy and the consequent glare. This policy is highly com-
mendable and the results of an adherence to this safe principle
are shown throughout the city.
An item of particular interest touched upon is the definition
of good service, which is given as perfect operation of the light-
ing appliances. In securing perfect operation the author has
aptly pointed out the necessity of not placing the entire burden
on the lamp manufacturer, but states in a truism that "The best
electric system is the man who runs it." The full meaning of
this attitude is perhaps appreciated most by those who have been
confronted with the operation side of lighting equipment, and a
good lamp coupled with the care thus implied in its every day
operation, is almost sure to result in satisfactory conditions for
all concerned.
If in the criticism of improvements in city street lighting, due
weight is placed on the necessity for moving slowly in the adop-
tion of untried apparatus, such criticism will be far less unreas-
212 TRANSACTIONS I. E. S. — PART II
onable than otherwise. It is surprising, however, even with
the conservatism this imposes, to note the many modern types
of lamps which are either in regular service or on trial in this
city. One thing which is nearly always noticed where lighting
improvements have been effected, is the raising of the standard
of illumination among even the unthinking. Thus the improve-
ments in the lighting of the avenues, has lead to criticism of the
side streets in Manhattan and the Bronx, and this in turn to
improvements of the side streets. This feature is often a great
help in extending higher and better illumination facilities.
The author has described the methods of engineering con-
nected with the new schemes of lighting. It is a cause of much
satisfaction to know that in the lighting of vast street areas, care-
ful attention is being given the question of illumination design.
Where the artistic side is naturally given so much weight, it
would be an easy matter to place rather more emphasis on this
feature than on the utility side.
The author has pointed out a principle of far reaching im-
portance when he states that the useful side must have as great
weight as the artistic, stating at the same time that it is equally
important not to cling too closely to the highest efficiency of the
units at the expense of an agreeable effect on the eye.
Throughout the whole address the author has shown that
lighting, while not a new question, is made up of many items
which are to-day looked at in new ways and which are solved
by new methods.
This is the key-note in the progress of illumination at this
time, and the fact that many of the items which concern the
final excellence of any lighting system are often simple and
even commonplace, should not under any circumstances lead
us to overlook their great importance to the results obtainable.
Mr. J. W. CowLES (communicated) : Mr. Lacombe sets forth
most interestingly the good results to be accomplished by the
adoption of a broad and systematic scheme for street lighting
under the varying requirements which exist in evCry city, and
the New York situation is a striking example of what can be
accomplished by close co-operation between the many interests
involved.
STREET LIGHTING OF NEW YORK 21 3
In many cities there may be seen street lighting apparently
developed with the one idea of illumination or utility in mind,
with practically no thought given to the artistic features, which
are certainly of value even though secondary to practical utility.
In other cases the reverse extremes are to be noted, but in New
York there is a striking and pleasing balance between both the
useful and the artistic.
I believe that much benefit can accrue to other municipalities
and public utilities by careful consideration of the points empha-
sized in this paper.
214 TRANSACTIONS I. Z. S. — PART II
ILLUMINATION OF PASSENGER CARS.*
BY J. Iv. MINICK.
Synopsis: This paper presents a brief record of the developments
in the methods of passenger car lighting that have been witnessed since
1825 when the candle was the source of light employed. Various types
of oil, gas and electric lamps which have been in general use are described
and illustrated. Illumination readings and data obtained from cars lighted
with lamps of the latter types are also included.
LIGHT SOURCES.
In their paper before the American Society of Mechanical
Engineers last winter, Messrs. Wood and Currie divided the
development of passenger car lighting into four twenty-five year
periods, beginning with the candle period in 1825. Oil lamps
came into general use about 1850 and gas about 1875. About
1900 electricity came into use as a means of lighting passenger
cars in steam train service, though it had previously been used
extensively in electric cars.
Information concerning the early use of candles is very meagre.
It is known, however, that Thomas Dixon, the driver of the first
passenger car, furnished his patrons with candles. He also
furnished a rough board table in the center of the car for sup-
porting the candles. The passengers were required to light the
candles and tend their feeble flames. The board table later gave
way to sockets attached to the walls, and these were superseded
by fixtures having glass chimneys to protect the flame and a coil
spring in the bottom of the socket to force the candle upward as
it burned away, thus maintaining the flame at a predetermined
position. This type of candle fixture is used extensively to-day
as an emergency lamp in case of failure of the primary gas or
electric system.
"Center-lamps," with one or more candles, came into use dur-
ing the latter part of the candle period. Many labor saving con-
veniences were developed, as for instance, an adjustable top to
hold the chimney in position without the aid of thumb screws,
* A paper read at a meeting of the Pittsburgh section of the Illuminating Engineer-
ing Society, May 16, 1913.
Fig. i. — Center deck candle fixture. About 1^40.
Fig. 2.— Center deck oil fixture. About iv.o.
Fig. 3.— Center deck gas fixture. About 1SS0.
Fig. 4. — Center deck electric fixture. About 19
minick: illumination of passenger cars 215
and brackets, that permitted of shifting the position of the lamp
both vertically and horizontally.
While comparatively little has been written concerning the
earlier types of oil lamps it is safe to assume that they resembled
the candle lamp in general design. It was comparatively inex-
pensive and quite convenient to remodel the candle fixture to
support an oil lamp. Such changes were very desirable as it is
a very difficult matter in car work to patch a hole in the side or
roof of a car without showing the patch.
Burners, wicks, etc., were adapted to the kind of oil used. The
use of the heavier oils, such as rape seed and Colza vegetable
oils, lead to the development of central draft burners, in which a
current of air was delivered to both sides of the flame to produce
more rapid combustion. The Argand, Belgian, acme and student
lamps are representative types of central draft burners. Two
wicks feeding one flame was another means of securing more
rapid combustion, and consequently a brighter light.
Many of the oil fixtures were equipped with telescoping attach-
ments for lowering the lamps for cleaning and filling. Others
had removable oil reservoirs. Reflectors came into use during
the oil period. In smoking cars, baggage and mail compartments
cheap metal or mirror glass disks were placed back of the lamp to
throw the light out into the car. In coaches conical opal glass
shades were slipped over the chimney and were supported by the
fixture arms.
Coal gas was probably the first kind of gas used in lighting rail-
way cars. It was secured from the city gas mains and stored
in a canvas reservoir, reinforced by wooden hoops, in the guards
van. Iron pipes, and rubber hose between the cars, served to
connect the lamps to the reservoir. Gasoline mixed with air was
very extensively used. Acetylene gas was also used to some ex-
tent. Pintsch gas, invented in 1867, came into very general use
on account of its reliability and increased storage capacity,
obtained by charging at high pressure.
A variety of burners were used, the first- of which was prob-
ably the flat or "fish tail" flame. There was also a central draft
burner somewhat similar to the central draft oil burner. The
substitution of mantles greatly improved the quality of the light.
2l6 TRANSACTIONS I. £. S. — PART II
For reasons previously explained the earlier types of fixtures
resembled the oil fixture. Generally the fixtures, used prior to
about 1905, were very ornamental in design to correspond with
the interior finish of the car. The introduction of the steel car
has changed this condition and present day fixtures have been
greatly simplified.
While it had previously been used in electric cars, the in-
candescent lamp did not come into general use as a means of
lighting passenger cars in steam train service until about 1900.
Carbon, metallized carbon, tantalum and tungsten filaments were
all used in about the order named, the latter type being in gen-
eral use to-day. It was the high efficiency of the tungsten lamp
that made electric car lighting possible, as the demand upon the
battery for current was brought within the necessary limits of
battery capacity and weight.
The earlier electric fixtures were generally gas fixtures re-
modeled to take incandescent lamps, many of which were not
equipped with reflectors. Such reflectors as were used served as
dust collectors, and by thus attracting attention, served to
stimulate the cleaning of the car.
The wide dissemination of knowledge of illumination, and the
constant and earnest study of the problems of serving the travel-
ing public, has resulted in better fixture designs, better distribution
of light, higher intensities, and higher efficiencies. Filigree work
has almost entirely disappeared. Simplicity of design has very
materially decreased initial costs and the use of reflectors specially
adapted to car service has made it possible to conceal the in-
candescent filament without the use of opal dipped or frosted
lamps. It should not be assumed, however, that the last word
has been said on this subject. There is a wonderful field for
further development and improvement.
ILLUMINATION.
While close attention is now being given to the proper lighting
of passenger cars, the chief effort, until within comparatively
recent years, was to reduce energy consumption and simplify
methods of operation. Lack of attention to the proper shielding
of the filaments, the better distribution of light, and the produc-
tion of intensities sufficient for the comfort of passengers, was
minick: illumination of passenger cars
217
largely due to lack of knowledge of this subject and lack of
facilities for accurately determining the conditions that obtained.
The development of the candle-foot photometer and other devices
has made it possible to determine all of these items and wonder-
ful improvements have been made in recent years.
The data herewith has been selected from a series of tests of
oil, gas, and electrically lighted cars, conducted during the past
three years. The cars were all taken from regular service °.nd
the results are therefore, representative of service conditions.
While the dimensions of the cars, the spacing and height of fix-
CAR
- X7^ °ES
ILLUMINATION
FOOT-CANDLE »:« N6S
Class P.f.
System Oil
Test Stan
1 -.-.
1 .64 Ha
61 lw .54
No 3219
Fixture 2Lt.Cen.Deck
Sead-na Ploir.e
.i'
j
.60 po
56;2w .50
Type *"Ocd
No r<»*ures 4
Max. F.C
.b4
3
.64 3a
52 .''w .42
Floor Area. Sol Ft. 367
Spacing iw**ire>V
(-1 in. F.C.
.24
s
•64,4a
61 4* .55
r n ;-
-. Ihl 95'
Aisle Av.F.t.
.5.-,
5
.61 ,5a
575k 53
UoDer Deck Liojht6reen
Rt«:ec*or CI.Ch^LOpSr.'iM
• Seats A». F.C.
.30
6
.53 5o
45 ;w 25
lower
LamD MO-SDua'. Burner
NirxJow ■ ■ »
.«4
7
.4)1 ;7d
33. 7w .35
Above Belt • Oak
Rating
Car Ay. r. C-
.48
8
.54 si
29 9w .24
5elew -
Lu"-!"S Jtil'.zed
114
. i ■ • -■ c n " . j 1 n
s-<? Lvr •:.-,?
E-H ciBncv
: '♦•'•!' [|
"1 — I — r
Fig. 5 —Two-light dual burner 6xtures— standing.
tures, color of interior finish, etc., vary to a slight degree, the
variations are not so great that comparison of the several types
of equipment cannot be made.
No special attention was given to the cleaning of the cars other
than to see that the lamps were cleaned and in proper adjust-
ment. All tests were conducted at night 'and the blinds were
drawn to prevent the leakage of outside light. When air draft
and temperature conditions tended to affect the value of the
illuminant. both standing and running tests were made. All
2l8
TRANSACTIONS I. E. S. — PART II
CAR
FIXTURES
ILLUHIK ATIOH
FOOT-CANDLE READING!
CIOJ5 Pf
Jvatem. Oil
Test. Runn
no.
'1
1.54
la 1-27 lw. US
Fintunt. 2 Lt C.ntcr deck
Readinc" -plane.
..36"
2
1.29 2a l-33|2w l>20
No. fixtures 4
Maximum F. C. "
1 44
J
I.I 1 pi M4 Jw ,.15
Floor Area Sq.rt 400.
■Spocinq I26"
Minimum F. C.
.62
4
14414a I.i3[4w 1. is
FlNI 5H
Heiqht 92"
Aisle M. F. C.
J
1.29 ja 1.25 law 1.02
Upper DecK Olive Green
RellectOr CI Chitniry or dom«
Aisle Seats F C.
1. 1 J
t
86' 6o .96 c •* ■ 92
Above Belt Liqht Oak
Rot ma
Car Av. F. C.
1 06
a
t^L
Ba ■ 71 law 62
Below •
Lumens utilised
4 52
3eolo Red riosh
Total lumens
Efficiency
UUUUUULJ14LILJLJLJ ,
!
#^
r=«-
"S?
<&
" f*^
l
Fig. 6. — Two-light burner fixtures — running.
CUR
FUTURES
ILLUMINATION
FOOT CANDLE F.EAOINES
Glass.
Pf
Sv3em . Oil
Test, Standing
1 125
la
111 llw
9E
No.
BJf.RH 3H0
Future. lit. Center Deck
Rearfina plane
36'
it
113
?a
LtOlcW
105
!yr*.
Wood
Number futures. 4
Ma.imum fc
UJS
3
■SA
3a
LOG
3W .99
Floor flruisn.ft.
400
Spocir.a,. 126'
Minimum fc
•SO
4
US
4a
at
Aw
1.0E
Finish.
Heiaht. qr
Aisle av fc
LOS
5
lei
5a
U8
5w .97
UppcrdetK .
Olivt qrten
Reflector CI rbimney- op.doim
Aisle scats a* F c
LOS
b
■SS
6a
•a
= w «5
L.'vrcrdeci.
Lamp. Acme burner
Window seats avfc
■31
7
.95
7a
«i
7w .79
Above tin.
Unfit Oak
Rating.
Carav.fc.
8
48
8a
•72
5w .60
Below bttti •
Lumens per lamp.
Lumens ulililtot
;su
Stat).
Ktd plush
Total lumens.
fffKiencv
Fig. 7. — Two-light burner fixtures— standing
minick: illumination of passenger cars
219
nOOT-CAHMJ PE»D:sSS
ILLUHIMATIQN
3 Bwlmi fgBJBBSfltf &?Ai!
jfe r. •_■<!£
'*.-■■'-'*
f'jr'
■ IS
*4.5c.
Lio)ht OoTT
Lamp' ln»trttal Artjanot Bun
1^*1
Window SecTfc Ay Ft.C
JO Ta .M7» .30
Sfflgjffiaiflim
E-frTuency "
Fig. 8. — Inverted Argand burner— running.
CAR
FUTURES
ILLUMINATION
FOOT CANOLF. RFAflNtS
Uau
S>iH" (.i.VL-r*(r &JMlmi
Tert Standing
1 1>»
U Lit
V .»
No WJ
ll»H 51U
Futures 1 Lt CcnTtr deck
Read^q plan! 3e"
t •»
l« .67
1» 7*
Wood
No f.ilom 5
M»» I.e.. 155
3 .74
3. .10
3« t5
F't.f 0r(| 1| ft
»!
Sp.t.na »•'
Hin ft -3»
4 n
«« -7«
tw M
nnuh
Hf.qh! "W
. . . .. • i •»
S 115
So US
5« M
. -< #. ■
Ol'wt qrifn
Rldfdor CI. Ajobt
A |1| ^.^» .». < < .10
i -78
S« .71
6» M
LI«H*C4
Lamp imtrtid Ara*i«* Vomer
*T«o™ stall av f.c .71
7 • »»
7a .»
7" ■»
«t..l [< •
r- :*•
Rot*,
Cor « re .78
v .... •
Lomtns utililM 116
'.J,
St.* 0 jiS
'.-3' .-(-A
tfF,c;tntr
UUUUUUUUUUUU ^
riifjiTfi ilifnf M
Fig. 9. — One-light inverted Argand burner— standing.
220
TRANSACTIONS I. E. S. — PART II
running tests were made while the cars were running at a uni-
form speed of about 40 miles per hour. So far as possible, all
equipment was adjusted to operate at the manufacturers rating
and where such adjustment was not possible proper correction
has been allowed. In the case of running tests, lamps were ad-
justed for running conditions and no change in adjustment was
made for standing tests. Foot-candle readings were taken on the
horizontal plane 36 (0.914 m.) inches above the floor.
Figure 5 shows a standing test of dual burner oil lamps.
This type of fixture consists of two lamps spaced on about 16-
inch (40.64 cm.) centers cross-wise of the car. Each lamp has
Sins.
Floor Area, Set Ft
5a. 1
HE
IhiSH.,
Upper Deck- Light Oli/eSreon
ZSaKI
Fixtures Z it. Center Deck
gp?c,lnq
Height
^effector CI.GI.'iVo o>6" 'in
tamp "_ Ihy, Am«nd Burner.
• ■,.■ ■
*.<-P..
kymsm per Lamp _Js
T°tal Wnena . r ifi
ILLUMINATION
Reading Plane
.-.'ar-ling I 1.02
Alsl» Av. F.C.
BEES
\y.r!\^V 'VriViz'e.'d'
S21
msney
v ;
FOOT-CANDLE READIN65
WIS
5i I I I Mi
<"*
a
z3
o
►"»
o'
o
k.
Fig. 10. — Two-light inverted Argand burner— standing.
two wicks feeding a single flame. Oil is fed through tubes from
a common central reservoir. Each lamp was equipped with clear
chimney and conical opal glass reflectors. Burners were adjusted
so that the center line of the flame was at right angles to the axis
of the car.
Figures 6 and 7 show running and standing tests respectively
with Acme burner oil lamps. This is a central draft type of
burner. Each lamp was equipped with a clear glass chimney, and
an opal reflector. The illumination was pleasing and sufficient for
newspaper reading. The light interior finish added to the appear-
ance of the car.
minick: illumination of passenglr cars
221
r00T-C*flPU PUP1HG5
1 Li Center Deck
l%h
lofci n
rmi3
Upper Deck Pea Grgg n
Height.
rroiV_i Gi^rVv^v
APo<« Pel' l.iq>it Oak
pjas:
f-o-vnfluih
tr— U UUUUUU4L444L4LJU,
ijiiy'ftiitiiifflifin'lF^
5m i I ||
4 L — .
J
Fig. ii.— One-light single mantle — running.
CIS
FUTU
RES
ILLUHINATI
□ N
. FOOr-CAHOLE KEADlNtS
On
SysUi
Pints* C«s
Tut
btandinq
l V |Ui tS8
lw 1*
Ho
PR 0 • X>9
Fotures
I Lt Center dec*.
Reoctna plane
36-
2 220
20 2J7
tw tS
Trr»
No Futures
rtarnnumfc
7.9S
3 225
3a 2J0
!w 187
floOfa'ca soft
i«
Spacma, 1061
» 122
4 2JI
' i yi
•» r
Finish
Heiqht
95
Aisle or fc
!2«
s m
ia m
SW «
ppvrAd
reaqreei
Reflector
FmtulGMeOI !
Aisle scats or fe
22Z
6 I9t
(a ■..«.«
6w 1>S
mi fled
Lamp
Laroe Mantle 3M4
window seats a* fc
04
I It:
7a Vl«
7* 5s
AW«e bill
UqMOpk
Reettiq
»e s n s c p
Carav fc
2(4
8 131
Sa 01
!W 112
Below belt
Lumens per Lamp
Lumens utilities
Joe
Seats
Brown Hull
Total Ihiis
3MS
Efficiency
.Us*
r-^JUUUUUUUUUUUUM
m m mrp m rfi m ffim
nl
r
1
Fig. 12— One-light single mantle— standing.
222
TRANSACTIONS I. E,. S. — PART II
ILLUMINATION
SJS55
FOOT- CANDLE READINGS"
'ypf
fixture I Lt Center Deck.
No.Fixturw
Reqd ir\c^ Plane
K.,, F.C
g 2.«2l2» 2.7^|gw 1.76
rc
27SJX1 2.4l|wi
'■!••/■
t'Beit UnjjrSTtfSJliijhrOS^
L nmp 50 w £:> luo.plen
fi,'4>Le ■Jeais a* fC £.35
5 2:76:Ja 2,5J,i>' ' 7^"
9 JL.4%1<\ 2.,'',".
y u lu u u u u u. u u u u u u g u y lj Lj ua u ii__j
Fig. 13. — One-light electric fixture— standing.
a
CAR
FIXTURES
ILLUMINATION
FOOT CANDLE READINGS
CI065 P-70
ElectnQ
'est Stondinq
la 1>3
iw W
No.
Fixture
' t Cc-'ter Beck
kcaui.nq plane 36
Z 333
2a Ml
lv| 1.5)
Type S'ee!
No FiAtur
5 10
Mat. f.c ).8t
in \3C
>tv .36
Floor Area, td. Ft. 538
Scccinfl
Min. f-c. Ill
4 Ml
4a 339
4« w
FINISH
Heiaht
Aisle o».U. 3.S7
5 W5
5q V5
'* :m
Reflector \UU S.F.
).'.',■ stitsa.U 126
£ lis
"5a" lib
*.-. nJ
Lamp
; . 3>
if. ;»o
Car av. t.c. 307
3 *;
s a 3;:
Bn. «S
Below » Dark Gr.cn
lomo 409
Lumens utilized 1652
1 3.31
9 a Ml
i» ni
Seats Green Flush
Wlumei
•; 40TO
EttlClMlcY «,«%
U 111 U U U U U U U U U LTD Lj D U U jJD U
Fig. 14.— One-light electric fixture — standing
I
minick: illumination of passenger cars 223
Figures 8 and 9 show running and standing tests re-
spectively of central draft types of carburetted gasoline equip-
ment. The burner is inverted in a clear glass bowl. Air from
the train system is passed through a spiral tube containing a wick-
soaked in gasoline, and the resultant mixture passes to the lamp
to be burned. The carburetter is placed on the top of the car
above the lamp so that the heat of the air raising from the lamp
will heat the incoming gas. This type of lamp is very susceptible
to changes of temperature and draft, and therefore requires con-
stant attention to prevent smoking, especially during stops at
stations.
Figure 10 shows the results obtained by the use of Pintsch
gas with inverted Argand burner lamps.
Figures 11 and 12 show running and standing tests re-
spectively of single mantle Pintsch gas lamps. Gas is carried in
tanks under the car at a pressure of about 150 to 160 pounds per
square inch (6.45 sq. cm.) and is reduced at the lamp to about
two pounds per square inch (6.45 sq. cm.) "With temperatures
below 20 deg. F., some of the hydro-carbons are precipitated with
a consequent reduction in illumination. It will be noted that
there is practically no difference in illumination between running
and standing conditions.
Figures 13 and 14 show the illumination that was obtained
in steel cars equipped with 50-watt. 60-volt tungsten filament
lamps, the first with flat prismatic reflectors and opal dipped
lamps and the latter with satin finish prismatic bowl type
reflectors with clear lamps.
CONCLUSIONS.
From these tests some idea of the comparative values of the
several types of car lighting units may be had. Similar tests of
cars varying only in interior color and finish will show changes in
efficiency of 100 per cent, or more, and other tests of direct, semi-
indirect and indirect fixtures will show considerable variation in
current consumption. The standard car lighting battery is none
too large from a capacity stand-point, while from a weight stand-
point it is now as large as it can be made for convenience in
handling. Changes in lamp efficiencies will be of little value until
224 TRANSACTIONS I. E. S. — PART II
they can be increased sufficiently to either reduce the weight or
increase the hours of battery service from 25 to 50 per cent.
Fixtures, reflectors and color of interior finish should receive
further attention with a view of reducing maintenance and
operating costs and providing better distribution of light at higher
efficiencies. Comparative tests should be made with side light
semi-indirect and indirect fixtures to determine their value not
only from a standpoint of illumination, but from the more im-
portant stand-point of current consumption.
DISCUSSION.
Mr. P. S. Millar: I have enjoyed Mr. Minick's talk very
much. His sketch of the history of the development of railway
cars is very interesting, while that part of it which pertains to
car lighting, and which is included in the printed paper, sets
forth a record of the improvement in this phase of illumination
which heretofore has been lacking in our Transactions.
In intensity of illumination the following is recorded in the
way of improvement:
Oil — flat wick 0.57 foot-candle
Oil — center draft; gasoline and Pintsch gas. .0.86 to 1.04
Pintsch mantle 1.99
Electric 2.60
But it is not alone in intensity that improvement has been
effected. The illustrations show that in the latest installations
with electric lamps, the filaments are shielded from ordinary view
and depolished reflectors are used. These installations show the
result of attention to car lighting which is very gratifying to all
of us who have to use the trains and who have in the past
suffered not only from inadequate lighting but also from bajl
lighting.
Recently I made a hurried trip from New ork to Boston and
return. My seat was in a car illuminated by a large number of
small tantalum lamps. The decoration of the car was dark so
that not only the ordinary glare from the lamps led to discom-
fort, but this was enhanced by the great contrast of the lamps
against the dark background. On the return journey I was
accompanied by a member of this Society who is prejudiced
against indirect lighting. It happened that we traveled in one
of the new trains on the New Haven road, which is equipped
ILLUMINATION OF PASSENGER CABS 225
throughout with indirect lighting fixtures. After a casual obser-
vation, my companion commented adversely upon the illumina-
tion, speaking of inadequacy of light, cheerless appearance, etc.
After five hours in the train, however, both of us were ready to
say that, ocularly speaking, we had never had a more comfort-
able railway journey. The single feature of concealed light
sources which characterizes indirect lighting was a most pleas-
urable element of lighting of a class in which exposure of light
sources is all too often the most prominent feature.
All things taken into consideration, I think we may well feel
gratified at the improvements in car lighting which are being
effected by the leading railroads, of which the installation just
referred to is one example.
Mr. A. C. Cotton : There are three general methods that
are or have been in use to produce energy for illuminating elec-
trically lighted cars. The first method was by the use of a
straight storage battery which had its difficulties on account of
the lamps sometimes being put on the battery while it was being
charged, the lamps thus receiving a much higher voltage with
a consequent shortening of the life of the lamps. When, there-
fore, the voltage dropped off to 1.8 volts per cell, the illumina-
tion was rather poor. Another method was by the use of a
generator driven by a steam engine or turbine placed in the
baggage car next to the locomotive, steam being furnished to
these units through a hose connection from the. locomotive. In
some cases the turbine was placed directly on the locomotive.
At times when there was a failure of steam on the locomotive,
the engineer either throttled the steam to the generator unit, or
turned it off altogether, which caused a failure of the electric
iights on the entire train. One great trouble with this system
was that in the event of the necessity of taking any one of the
electrically equipped cars from the train, and substituting there-
for either an oil or gas lighted car, all electrical lighted cars
back of this car were in darkness. This caused considerable
discomfort to the passengers and the railroad companies were
severely criticised. On account of this and other difficulties,
the head-end systems, whereby the cars were illuminated from
a common train-line without the use of storage batteries, have
been quite generally discarded.
226 TRANSACTIONS I. E. S. — PART II
The system quite generally used to-day is that known as the
axle generator type, wherein each car is equipped with a gene-
rator suspended from the truck beneath the car, and driven by
a pulley fastened to one of the axles. Each car is also equipped
with a storage battery, generator regulator, for controlling the
generator, and a lamp regulator, for controlling the voltage of
the lamps. There are several different types of axle generator
systems in use to-day, both foreign and domestic, but the ones
we are principally interested in, are those manufactured in this
country, as the foreign types have gained very slight foothold
up to the present time. The generator in this system is equipped
with a pole-changer, so that the polarity of the wiring is always
the same, irrespective of the direction of the movement of the
train. The generator regulator governs the point at which the
generator is thrown in on the battery and also governs the output
of the generator by changing the strength of the field of the
same. The lamp regulator, which may be either motor operated
or magnetically operated, acts so as to keep the voltage of the
lamps constant, irrespective of the voltage of the generator or
battery. By means of this system, the storage batteries on the
cars are automatically charged while the train is in motion, the
generator supplying current to either the battery or lamps, or
both, while, when the train is standing, the current is supplied
to the lamps from the battery.
Mr. J. L,. Minick (In reply) : The use of electricity, as a
means of lighting passenger cars on the Pennsylvania Railroad,
was brought about largely by reason of the improvements in
the vicinity of New York City. On account of the large amount
of trackage in tunnels under the North and East Rivers and
Manhattan Island, it was thought wise to use only such equip-
ment as contained no inflammable materials of any kind, conse-
quently electricity became the agent for supplying light. As
practically all of the steel equipment is likely to enter New York,
electric light has become the prevailing system.
The length of run on a single battery charge, will of course,
depend upon the demand for energy during the run. The bat-
teries on the cars of. the Pennsylvania Railroad have a rated
capacity of 300 ampere-hours. Some of the express cars have
a maximum current demand as low as 2.4 amperes, while certain
1 1.]. I'M I NATION OF PASSENGER CARS 227
dining cars may exceed 45 amperes. So that the length of run
may vary from 125 hours to about 6.5 hours of lighting. Coaches
with 63-volt equipment have a current demand of about 10 am-
peres. Coaches have been run from New York City to St.
Louis on one charge. I am not prepared to give the length of
run possible on a full gas charge.
Referring to Mr. Millar's remarks, I wish to say that from
the standpoint of illumination alone, I am very favorably in-
clined towards some form of indirect lighting. In car work,
however, there are many considerations in addition to that of
proper illumination. Standards have been fixed at an average
intensity of three foot-candles for coaches and about five foot-
candles for dining cars. The size and weight of the present
300 ampere-hour battery is as great as it can possibly be made
for convenience in handling, while from a current capacity stand-
point it is none too large. If semi-indirect lighting be used the
current demand will probably be increased by 100 per cent, for
the same intensity and if totally indirect be used it may be in-
creased by several hundred per cent. If it be attempted to main-
tain equal brilliancy in the car the current demand will again
be increased several times. Under present conditions I do not
believe it possible to provide either semi-indirect or totally in-
direct lighting at a reasonable demand for energy.
When berth lamps were first used they were a decided im-
provement over the then existing conditions. I understand that
an effort is now being made to entirely conceal the lamp but
the space available in existing cars is probably so limited that
changes for the better cannot be made conveniently. In the
lighting of trolley cars it is necessary to place several lamps
in series across the circuit on account of the high voltage used.
As the failure of any lamp in this series puts the entire series out
of commission it is necessary to provide several series circuits
for a single car. In a small car this means a large number of
small lamps. If the total number of lamps could be reduced I
have no doubt more consideration would be given to correct
lighting.
Mr. L. C. Porter: There are two sentences on the seventh
page of Mr. Minick's paper, to which I would like to call atten-
tion. They are as follows : "The illumination was pleasing and
228 TRANSACTIONS I. E. S. PART II
sufficient for newspaper reading. The light interior finish added
to the appearance of the car." It seems to me that there should
be a great deal more attention paid to the esthetic effects of the
interior finish of cars, than is at present. If a car is finished
entirely in dark color, no amount of light will make it appear
bright and cheerful.
Photometer tests, while useful in comparing the actual effi-
ciency of utilization of different systems, do not tell the complete
story, and should be supplemented by personal observation.
Mr. Minick has asked for a discussion of the plane on which
illumination measurements should be made in railway cars, call-
ing attention to the fact that some people are making them on a
45 deg. plane, which is the plane in which a reader would natu-
rally hold a paper, while others are making them on the horizon-
tal plane. I believe both planes should be used. It is necessary
to find the average illumination on the horizontal plane, in order
to calculate the effective lumens and thus determine the efficiency
of utilization of the lighting system, interior finish of the car, etc.
Measuring on the horizontal plane, however, will not take into
account shadow effects.
With high-power units and wide spacing, a passenger at the
end of the car, seated with the first illuminant a considerable dis-
tance in back of him, might receive good light on his paper, while
on the other hand, if he were facing the lighting unit, the side
of the paper towards him would be in shadow.
I believe that at each station readings should be taken on three
planes, namely, the horizontal, the 45 deg. towards one end of
the car, and the 45 deg. plane towards the other. Multiplying
the horizontal reading by the ratio of the two 45 deg. planes,
always dividing the low reading by the high, will give a figure
which will show more nearly the effectiveness of the illumination.
For example, suppose the intensities on these three planes with
two different lighting systems were as follows :
Foot-candles Foot-candles Foot-candles
Horizontal plane 45 deg. rear 45 deg. front
Sj'stem A 3 3. 20 2.56
System B 3 3.20 1.60
2. 56
Then from system A" we have 3 X — — = 2.4, and from B we
3.20
get 3 X — — = 1.5. Clearly system A is the better system.
TRANSACTIONS
OF THE
Illuminating Engineering Society
Published monthly, except during July, August, and September, by the
ILLUMINATING ENGINEERING SOCIETY
General Offices: 29 West Thirty-Ninth Street. New York
Vol. VIII
JUNE, 1913
No. 6
Council Notes.
A regular meeting of the Council was
held in the general offices of the society,
29 West 39th Street, New York, June
13, 1913. Those present were: Preston
S. Millar, president; Charles 0. Bond,
George S. Barrows, L. B. Marks, Nor-
man Macbeth, Joseph D. Israel, general
secretary, W. J. Serrill and George H.
Stickney by an invitation.
A monthly report on the finances and
membership of the society was re-
ceived from the general secretary.
Upon recommendation of the Finance
Committee, bills aggregating $1,833.99
were authorized paid.
Twelve applications for membership
and four resignations were accepted.
Counting these changes, the member-
ship was said to total 1,373 members.
Reports of progress were received
from the following committees : Recip-
rocal Relations with Other Societies,
Collegiate Education, Advertising, No-
menclature and Standards, Progress,
Research and Glare.
A report was received from the
Committee on Tellers giving the results
of the election of officers for the
society and the several sections. The
names of the officers elected appear
elsewhere in this issue.
A report of progress on the work
of the Philadelphia Section was re-
ceived from Vice-President W. J.
Serrill.
Two applicants, The Consolidated
Gas, Electric Light & Power Company
of Baltimore and the Welsbach Com-
pany, were elected sustaining members.
Additional appointments by the presi-
dent to various committees were ap-
proved.
The general secretary was directed to
prepare a report on the work of the
Council during the present year to be
submitted to the membership of the
society.
The Executive Committee was em-
powered to act for the Council during
the summer months.
A communication from the Heights
of Buildings Committee of the City of
New York inviting the society to co-
operate with the committee in its work
was read. The president was directed
to acknowledge the communication and
to state that the society will be glad to
co-operate with the committee as far
as possible.
New Members.
The following applicants were elected
members of the society at a meeting of
the Council, June 13, 1913:
Butler, Henry Emanuel
Asst. in Illuminating Engineering
Laboratory, General Electric Com-
pany, Schenectady. N. Y.
TRANSACTIONS I. B. S. — PART I
English, J. C.
President, J. C. English Company,
128 Park Street, Portland, Oregon.
Freemen, E. H.
Professor of Electrical Engineering,
Armour Institute of Technology,
33rd Street & Armour Avenue, Chi-
cago, 111.
Johnson, Otis L.
Illuminating Engineer, Benjamin
Electric Mfg. Company, 120 So.
Sangamon Street, Chicago, 111.
Klingman, A. M.
Asst. Commercial Engineer, Na-
tional Quality Lamp Division of
General Electric Company, Nela
Park, Cleveland, O.
Langan, Joseph
Assistant Secretary, Illuminating
Engineering Society, 29 West 39th
Street, New York, N. Y.
Roland, E. U.
Headlight Field Man, Remy Electric
Company, Anderson, Ind.
Schott, Albert
Electrician, McCreery & Company,
6th Avenue & Wood Street, Pitts-
burgh, Pa.
Simpson, Richard E.
Asst. in Illuminating Engineering
Laboratory, General Electric Com-
pany, Schenectady, N. Y.
Strang, Perry S.
Inspector, National X-Ray Reflector
Company, 11 19 West Jackson Boule-
vard, Chicago, 111.
Wangersheim, E. A.
President, General Lighting Fixture
Company, 28 West Lake Street,
Chicago, 111.
Wood, Douglass
Illuminating Engineer, Bryan-Marsh
Electric Works, 431 So. Dearborn
Street, Chicago, 111.
Sustaining Members.
At a meeting of the Council, June 13,
1913, the Consolidated Gas, Electric
Light and Power Company of Baltimore
and the Welsbach Company were elected
sustaining members of the society.
Section Notes.
CHICAGO SECTION.
At a meeting of the Chicago Section
in the Auditorium of the Western So-
ciety of Engineers, June 27, Mr. Arthur
J. Sweet presented a paper entitled
"Notes on Postal Car Illumination."
The following officers have been
elected for the year beginning October
1, 1913: chairman, Dr. M. G. Lloyd;
secretary, J. B. Jackson; managers, J.
W. Pfeifer, Dr. Nelson M. Black, C. C.
Schiller, M. J. Sturm and H. B.
Wheeler. Mr. Jackson was re-elected
secretary.
NEW ENGLAND SECTION.
The following officers have been
elected for the ensuing year : chairman,
C. A. B. Halvorson, Jr.; secretary, H.
Harold Higbie ; managers, R. B. Hus-
sey, H. C. Jones, J. M. Riley, R. C.
Ware and W. E. Wickenden. Since the
election, Prof. Higbie has removed from
the territory of the New England and
has tendered his resignation as secretary.
A secretary will be appointed by the
new section board.
NEW YORK SECTION.
The following officers have been
elected for the ensuing year : chairman,
W. Cullen Morris; secretary, Clarence
L. Law; managers, H. B. Rogers, C. R.
Clifford, H. V. Allen, W. H. Spencer
and Oscar Fogg. Mr. Law was re-
elected secretary.
TRANSACTIONS I. E. S. — PART I
PHILADELPHIA SECTION.
The Philadelphia Section held a din-
ner at the Engineers' Club, 1317 Spruce
Street, Friday evening, June 20. The
dinner was held in place of the regular
June meeting and was attended by
twenty-five members. Prof. James
Barnes of Bryn Mawr College gave a
short talk on "Recent Ideas in Regard
to the Spectrum." Dr. Wendell Reber
gave a short address on "The Problem
of the Relation of the Human Eye to
Illumination, Natural and Artificial."
The following officers have been
elected for the year beginning October
1 : chairman, Prof. George A. Hoadley ;
secretary, L. B. Eichengreen ; managers,
F. C. Dickey, H. H. Ganzer, H. A.
Hornor, H. Calvert and Samuel Snyder.
Mr. Eichengreen was re-elected secre-
tary.
PITTSBURGH SECTION.
At a meeting of the Pittsburgh Sec-
tion in the auditorium of the Engineers'
Society of Western Pennsylvania,
Oliver Building, June 20, Messrs. Evan
J. Edwards and Ward Harrison of the
National Electric Lamp Association
presented a paper entitled "Some Engi-
neering Features of Office Lighting."
The paper will be published in the next
issue of the Transactions, the October
number.
Announcement was made of the elec-
tion of the following officers for the
ensuing year: chairman, C. J. Mundo ;
secretary, Alan Bright ; managers, H. S.
Hower, H. H. Magdsick, E. R. Roberts,
C. E. Stephens and S. B. Stewart.
October 1 : president, Charles O. Bond ;
vice-president to represent the New
York Section, George H. Stickney;
vice-president to represent the Pitts-
burgh Section, Ward Harrison ; general
secretary, Joseph D. Israel ; treasurer,
L. B. Marks ; directors, F. J. Rutledge,
C. A. Littlefield, F. A. Vaughn. Di-
rectors are each elected for three
years, vice-presidents, two years, and
the other officers one year. Messrs.
Israel and Marks were re-elected.
The results of the election for officers
of the several sections may be found
under the Section Notes in this issue.
New Officers.
At the recent annual election of the
Society the following officers were
elected for various terms beginning
Charles O. Bond, president-elect of
the Illuminating Engineering Society,
was born November 15, 1870, near the
town of Lehigh, Webster County, Iowa.
He was graduated from the United
States Naval Academy, Annapolis, Md.
in 1890. After graduation he served
one year at sea on board the U. S. S.
Enterprise and U. S. S. Philadelphia,
resigning from the navy in 1891. He
later taught school in the states of
Iowa and New York for five years, and
in 1897 became connected with the
United Gas Improvement Company in
Philadelphia. In 1898, during the
Spanish-American War, he served five
months as an ensign in the navy on
board the U. S. S. Lancaster and the
U. S. S. Newport. At the close of the
war, he resumed his connection with the
United Gas Improvement Company, tak-
ing charge of the photometric work of
the company. While continuing in this
position, he also held command of one
division of "the Naval Force of Pennsyl-
vania for three years. Since 1909, Mr.
Bond has been manager of the photo-
metrical laboratory of the United Gas
Improvement Company, which was es-
tablished in that year.
4
TRANSACTIONS I. E. S. — PART I
Joseph D. Israel, who has twice been
elected general secretary of the society,
has been connected with the lighting
industry during the past twenty-six
years. He is at present district manager
of the Philadelphia Electric Company.
Mr. Israel was born in Philadelphia,
February 28, 1868. In 1886 he was
graduated from the Scientific School of
the University of Pennsylvania with the
degree of Bachelor of Science. After
a post-graduate year in mechanical and
electrical courses, the same university
conferred upon him the degree of Me-
chanical Engineer in 1887. Immediately
after graduation he became connected
with the Edison Electric Light Com-
pany of Philadelphia, devoting his time
to underground street work. Shortly
afterward he became superintendent of
the street work of the company. He
next became assistant to the manager
of the company, and later was made
secretary and manager. When the com-
pany was merged with the Philadelphia
Electric Company he became district
manager of the latter company.
Mr. Israel has been active in the local
and national work of the Illuminating-
Engineering Society, the National Elec-
tric Light Association and the American
Institute of Electrical Engineers. He
is also a member of the Franklin Insti-
tute of the State of Pennsylvania and a
director of the Commercial Section of
the National Electric Light Association.
He has contributed papers and reports
to local and national meetings of the
above-named societies, and the Associa-
tion of Edison Illuminating Companies.
A Survey of Present Day Lighting.
The Illuminating Engineering So-
ciety, through its president, Mr. Preston
S. Millar, is undertaking the preparation
of an exhaustive survey of present day
lighting conditions, the object being to
record as nearly as possible the lighting
practise in a number of different fields.
It is expected that the survey will af-
ford concrete information on the pres-
ent situation and provide a basis of
comparison for future estimates of
progress. It will probably be of most
value at the present time in enabling
individuals and companies to compare
their practises with general practise
and conditions ; it should disclose what
is judged the most advanced practise in
each field.
For the purpose of this survey the
United States has been divided into
eighty-nine different sections, each sec-
tion containing approximately one mil-
lion inhabitants. Into each district vari-
ous lists of questions regarding illumi-
nating practise are being sent to repre-
sentative companies and individuals in
the following professions and indus-
tries :
Central stations.
Gas companies.
Municipal engineers.
Manufacturers of incandescent lamps.
Manufacturers of mantle burner lamps.
Manufacturers of arc lamps.
Manufacturers of acetylene supplies, tips, etc.
Manufacturers of oil lamps.
Manufacturers of small isolated lighting plant
equipments, gasolene, acetylene, etc.
Manufacturers of lighting glassware.
Fixture manufacturers.
Arc lamp post manufacturers.
Ophthalmologists.
School associations and commissions.
Railroads.
Street railroad companies.
Street lighting lamp companies.
It is hoped, therefore, that with a
reasonable amount of co-operation on
the part of those to whom the
questions are sent that estimates and
"" C<.'
S
CHARLES O. BOND, President-elect.
JOSEPH D. ISRAEL, General Secretary.
\
TRANSACTIONS I. E. S. — PART I
data will be received from some two
to three thousand persons, each of
whom is probably best fitted to pro-
vide the information for which he is
asked. In interpreting and summariz-
ing the information, which it is hoped
will be made available, an effort will be
to emphasize that which is constructive
and to avoid everything invidious. The
final survey will be made up with the
co-operation and advise of a number of
men in the lighting industry, and no
information will be used which will be
likely to prove in any way derogatory
to the interests of those contributing.
The information derived will be used as
a basis of a paper to be presented at the
convention of the Illuminating Engi-
neering Society in Pittsburgh during
the week beginning September 22, 1913.
6 TRANSACTIONS I. E. S. — PART I
ANNOUNCEMENT BY COMMITTEE ON NOMENCLA-
TURE AND STANDARDS.
The committee has tentatively adopted the following addi-
tional definitions :
Apparent candle-power, at a distance d, is the candle-power of
the simple luminous source which at the distance d from the
point of observation would give an illumination equal to the ob-
served illumination.
The term should be used only for cases in which the law of
inverse squares does not apply. The term is meaningless unless
d is given.
Power consumed by a source, P, expressed in ergs per second is
the total power input in a radiating body.
Power radiated by a source, Pr = 1 PrA dK, the radiant
power emitted by a source in the form of radiation between wave-
lengths zero and infinity, expressed in ergs per second.
pr
Radiation Efficiency, p — — , a numeric, is the ratio of the
power radiated to the total power consumed by a source.
Specific Consumption, the ratio of the power consumed by a
source to the total luminous flux in lumens.
Specific luminous output, the reciprocal of specific consump-
tion, F/P.
The committee further would present to the membership the
following definitions proposed by Dr. H. E. Ives and not yet
acted upon by the committee. The committee desires the sug-
gestions and criticisms of the membership on all of the definitions
here given.
po.S/u
Available useful power for lighting purposes, Re = I RA d\ =
Jo.4fi
the radiation lying in the visible spectrum. (The power which
gives one mean spherical candle of this radiation is sometimes
called the mechanical equivalent of a light).
Visible fraction of the radiation from a source = ~~ = fraction
of the total radiation useful for lighting purposes. (A pure
TRANSACTIONS I. E. S. — PART I
numeric). (This fraction is sometimes called the radia?it luminous
efficiency, the ratio -e , the total luminous efficiency of a light
source).
R j. *
Radiant specific consumption = -=r = power radiated per
lumen = p-^r-
F
Radiant specific luminous output —=- = lumens radiated per
F
watt radiated = Km= -=■ (.*. identical with stimulus coefficient) .
p¥
Note:— Watts per candle, candles per watt, watts per lumen, lumens
per watt, are ratios sometimes called " efficiency."
Radiant Luminous Efficie?icy = fxR = ratio of the radiant
specific luminous output of a source to the maximum possible
specific luminous output. (A pure numeric)
JF
R Km
Pr
(i),
i rc
' max J
KA Ra a\.
Total luminous efficiency = /» = ratio of the specific luminous
output of a source to the maxium possible specific luminous
output. (A pure numeric)
P R _ K^p
K„
(f L (1)
max
max
Since maxium value of p is unity.
Note : — If Kmax be taken as unity, i. e., if the unit of flux be taken as
that given by one power unit of radiation of maxium luminous efficiency,
then considering the radiation from a source
F _
MR = -g- = Kg,
or luminous efficiency = specific luminous output = stimulus coefficient
(numerically).
For the total radiation
F _
or luminous efficiency is numerically the same as specific consumption.
C. H. Sharp,
Secretary Committee on Nomenclature and Standards.
3 TRANSACTIONS I. E. S. — PART I
Constitution and By-laws
OF THE
Illuminating Engineering Society
(Adopted by a vote of the Membership, January 14, 1907.)
Constitution amended: Jan., 1909; Jan., 1910; Jan., 1911; Jan., 1912; Dec,
1912. By-laws amended by Council: Jan. 28, 1907; Feb. 10, 1910;
Mar. 10, 1911 ; Dec. 8, 1911; Jan. 12, 1912; Jan. 10, 1913.
ARTICLE I.
NAME AND OBJECTS.
Section 1 : The name of this association shall be the Illumi-
nating Engineering Society.
Section 2: Its objects shall be the advancement of the theory
and practise of illuminating engineering and the dissemination
of knowledge relating thereto. Among the means to this end
shall be meetings for the presentation and discussion of appro-
priate papers ; the publication as may seem expedient of such
papers, of discussions and communications ; and through com-
mittees, the study of subjects relating to the science and art of
illumination, and the publication of reports thereon.
(a) Sec. 2: The appointment of committees to report upon
scientific and engineering subjects shall be authorized only by
a majority vote of the Council, which shall be taken by letter-
ballot. When such a committee is thus authorized, the President
shall appoint the members thereof, subject to approval by vote of
a quorum of the Council.
ARTICLE II.
MEMBERSHIP.
Section 1 : The members of this Society shall be designated
as Members, Sustaining Members and Honorary Members.
Section 2: A Member may be anyone interested in the objects
of the Society. At the time of his election he shall not be less
than twenty-one years of age.
By-laws s re printed in small type.
TRANSACTIONS I. E. S.— PART I 9
Section 3: A Sustaining Member may be a company, firm,
association or individual interested in the objects of the Society
and desirous of contributing to its support. A Sustaining Mem-
ber, when other than an individual, may be officially represented
by an individual. The privileges of Sustaining Members shall
be the same as those of members, except the right to vote and
to hold office. All provisions of this Constitution governing the
admission, duties and obligations of members shall, unless other-
wise provided, apply to Sustaining Members.
Section 4: Honorary Members may be chosen from among
those who are of acknowledged eminence in some branch of art
or science related to illuminating engineering. Honorary Mem-
bers shall be entitled to all the privileges of the Society except
the right to vote and to hold office therein.
ARTICLE III.
ADMISSION AND EXPULSION OF MEMBERS.
Section 1 : Honorary Members shall be proposed in writing
by at least fifteen members, and shall be elected only by the
unanimous vote of the Council. Voting shall be by letter-ballot.
A person elected an Honorary Member shall be promptly noti-
fied by letter, and the election shall be cancelled if an acceptance
is not received within six months after the mailing of such
notice.
Section 2: An application for admission to the Society shall
be made in a form prescribed by the Council and shall bear the
endorsement of at least two Members of the Society; or shall
refer to at least two Members of the Society; or if an applicant
certifies that he is not personally known to two members, refer-
ences may be accepted to members of professional societies of
good standing, or to other persons whose good standing may be
readily verified.
(a) Sec. 2: An application for membership in the Society
shall be made upon a printed form prepared by the General Secre-
tary and approved by the Council, which shall call for such in-
formation as may be required by the Board of Examiners and
the Council to pass properly upon the eligibility of a candidate.
By-laws are printed in small type.
10 TRANSACTIONS I. E. S. — PART I
(b) Sec. 2: In the absence of replies from referees to in-
quiries for information, or if replies are not sufficiently explicit,
the Board of Examiners having cognizance of the application
shall cause the applicant to be notified and shall hold his appli-
cation in abeyance.
Section 3: All applications for admission to membership shall
be passed upon by a Board of Examiners of the section of the
Society representing the locality in which the applicant resides.
All applications shall be reported to the Council for final action.
An applicant not residing within the territory of a section shall
submit his application direct to the Council.
(c) Sec. 3 : When applications for admission are received
from persons residing within the territory of a section the
General Secretary shall notify the Secretary of that section to
make prompt report upon the application.
(d) Sec. 3: The privileges attaching to membership in the
Society shall not be accorded to newly-elected members until
they have paid their entrance fee and current dues. This by-law
shall be printed upon the notification of election.
(e) Sec. 3: Upon receipt of an application for membership,
which shall be made on the official form the General Secretary,
or the Secretary of a section, shall see if it has been properly
filled out. If not, he shall return the form and notify the appli-
cant of the deficiency. When an application is in proper form
it shall be forwarded to the chairman of the Board of Examiners.
The Secretary of a section shall conduct for the Board of Ex-
aminers such correspondence with applicants and their referees
as the Board may direct.
(f) Sec. 3: Objection to the admission of a candidate must be
accompanied by specific reasons for such objection.
Section 4: A Member may resign from the Society by a
written communication to the Secretary, which resignation shall
be accepted by the Council if all his dues and other indebtedness
have been paid, and the Society badge has been returned.
Section 5: Upon the written request of ten or more members
that, for cause definitely stated in detail, a Member of the So-
ciety be expelled, the Board of Managers of /the section of the
locality wherein the accused resides shall consider the matter,
and if there appears to be sufficient cause shall advise the ac-
cused of the charges against him. The accused may then present
By-laws are printed in small type.
TRANSACTIONS I. E. S. — PART I 11
a written defense and appear in person before a meeting of the
board. The finding of the board shall then be submitted to the
Council of the Society which, within two months, shall finally
consider the case, and if a satisfactory defense has not been
made, the accused Member shall be expelled upon a two-thirds
vote of the Council. In the case of one not a member of a
section, charges shall be preferred directly to the Council.
ARTICLE IV.
DUES.
Section 1: An entrance fee, payable on admission to the
Society, may be fixed by the Council.
(a) Sec. i : The entrance fee for members shall be $2.50 pay-
able on admission to the Society. There shall be no entrance fee
for Sustaining Members
Section 2: The annual dues for members shall be $5, which
shall include subscription to the Transactions of the Society.
(b) Sec. 2: The annual dues are payable in advance. Bills
for dues shall be sent out by the General Secretary not later
than October 10.
(c) Sec. 2: If the entrance fee and dues are not paid within
one month after a member has been notified of his election, he
shall be finally informed of the delinquency; and if such dues
are not paid within two months from the time of notification of
election, the Council shall cancel the election, of which cancella-
tion the delinquent and his referees shall be informed. This by-
law shall be printed on the final notice above provided for. (See
also by-law Sec. 3, Art III.)
(d) Sec. 2: Any member in arrears four months for dues,
shall be informed by the General Secretary that he is delinquent
and can have no vote or voice in the affairs of the Society or
receive its Transactions or other publications until the dues are
paid. At the expiration of two months thereafter, if still in ar-
rears, he shall be notified that his name will be presented to the
Council as delinquent, if the dues are not paid within one month.
If the member continues delinquent, the Council shall drop him
from membership at the regular meeting held in June.
(e) Sec. 2: From the annual dues paid by each Member $3
shall be deducted and applied as a subscription to the Transac-
tions for the year covered by such payment. The price of sub-
By-laws are printed in small type.
12 TRANSACTIONS I. E. S.— PART I
scription of the Transactions to non-members of the Society
shall be $5 per year. Single copies may be sold at 55 cents each ;
provided that volumes reserved shall not be broken to furnish
single copies.
(f) Sec. 2: The official badge of the Society shall be issued
by the General Secretary upon application to any Member in
good standing, upon payment of $3 ; provided that Honorary Mem-
bers shall receive the badge without payment. Each badge shall
be numbered and registered in the name of the member receiving
it. Members purchasing badges shall be informed by the Gen-
eral Secretary that they are issued with the express condition
that if the member resigns or is dropped from the Roll of the So-
ciety, he shall return his badge, receiving therefor the sum of $2.
(g) Sec. 2 : A certificate of membership in the Society shall
be issued by the General Secretary upon application to any mem-
ber in good standing, upon payment of $1.00.
Section 3: The annual dues for Sustaining Members shall be
not more than $250.00.
(h) Sec. 3: Dues (not exceeding $250 annually) for Sustain-
ing Members may be elective with the Sustaining Member.
Transactions shall be sent free of charge to Sustaining Mem-
bers whose dues are $10.00 or more per annum.
Section 4: Honorary Members shall be exempt from all pay-
ments.
Section 5: A Member elected after six months of the fiscal
year have expired shall pay one-half of the amount of dues for
that year; provided, that if he requests and receives a set of
Transactions covering the entire year, then the full annual dues
shall be paid.
Section 6: A Member who has been dropped as delinquent
may be reinstated by the Council and retain his original date of
election upon payment of all back dues, being then entitled to a'
complete file of the publications of the Society, if in stock, cor-
responding to the period of delinquency.
ARTICLE V.
OFFICERS.
Section 1: The officers of the Society shall be a President.
Vice-presidents equal in number to the number of organized
Sections, nine Directors, a Secretary and a Treasurer.
By-laws are printed in small type.
TRANSACTIONS I. E. S. — PART I 13
Section 2: The President, the Secretary and the Treasurer
shall hold office for one year ; the Vice-presidents shall hold office
for two years and the Directors for three years. Terms of of-
fice shall commence the first day of October. A retiring Presi-
dent, Vice-president or Director shall not be eligible for imme-
diate re-election to the same office, and a retiring Vice-president
shall not be eligible for immediate election as a Director. At
each annual meeting officers shall be elected to succeed those
retiring by expiration of term.
Section 3: A vacancy in the office of President shall be filled
by the senior Vice-president ; a vacancy in the office of Vice-
president shall be filled by the senior Director ; a vacancy in the
office of Director shall be filled by the Council, preferably by
selection from members, if any, who at the previous annual
election received votes for the office of Director. A vacancy in
the office of Secretary or Treasurer shall be filled by the Council.
Such succession to office or appointment by the Council shall not
render an officer ineligible for immediate election to the same
office. Seniority between officers of the same rank and date of
election shall be determined by the date of their election as mem-
bers.
Section 4: No officer shall receive, directly or indirectly, any
salary, compensation or emolument from the Society, either as
such officer or in any other capacity, unless authorized by a vote
of the majority of the entire Council. No officer shall be inter-
ested, directly or indirectly, in any contract relating to the opera-
tions conducted by the Society, nor in any contract for furnishing
supplies thereto, unless by the unanimous vote of the Council.
ARTICLE VI.
ELECTION OF OFFICERS.
Section 1: Each year not later than April I, a Board of
Nomination, consisting of the two junior Past-presidents and of
the Past-vice-presidents whose terms of office expired in the two
preceding Septembers, shall proceed to prepare a nomination
ticket containing the names of those whom they deem best suited
for the offices to be filled at the ensuing annual election. Nomi-
2
14 TRANSACTIONS I. E. S. — PART I
nees for the office of Vice-president shall be so selected that, if
such nominees are elected, each locality where there is a Section
of the Society may be represented on the Council by a Vice-
president.
(a) Sec. i : Each year, not later than March 15, the General
Secretary shall send to the senior Past-Officer among those desig-
nated in Article VI, Section i, a copy of that section and the
names and addresses of the other Past-Officers therein designated.
The senior officer shall then forthwith proceed to organize the
Board of Nomination and shall submit its report to the General
Secretary not later than April 25.
Section 2: The ticket thus prepared shall be printed and for-
warded to members not later than May 5 together with an un-
marked inner envelope, and an outer official voting envelope
bearing the name and address of the Society and the words,
"Official Voting Envelope — Enclosing a Ballot Only." The
member voting shall enclose his ballot in the unmarked envelope,
which shall, in turn, be enclosed in the outer envelope, which
latter shall be endorsed with the name of the sender. Ballots to
be counted must reach the General Secretary not later than
May 26.
(b) Sec. 2 : The General Secretary shall have printed and
enclose with the official nomination ticket, Section 2 of Article
V, and Sections 1, 2 and 3 of Article VI.
(c) Sec. 2: The roll of members shall designate those who
are charter members of the Society. The names of present and
past general officers shall be followed by the name of office held,
printed in italic type.
Section 3: A member may vote the official ticket above pro-
vided for; or he may erase any names thereon and substitute
others; or he may substitute a written ballot containing names
of his own selection.
Section 4: The President at a Council meeting in May shall
appoint, subject to the approval of the Council, five members, not
members of the Council, to constitute a Committee of Tellers.
This committee shall meet between May 26 and May 30, and
shall receive unopened all ballots from the General Secretary and
shall forthwith proceed in secret to count the vote. It shall then
By-laws are printed in small type.
TRANSACTIONS I. E. S. — PART I 15
prepare in duplicate and sign a report of the results of the vote,
one copy of which shall be delivered to the General Secretary and
the other handed by the chairman of the committee, at the en-
suing annual meeting, to the presiding officer, who shall at the
opening session of the meeting announce the names of the
officers elected.
(d) Sec. I : The General Secretary upon receipt of the report
from the Committee of Tellers, shall at once notify the president-
elect of the result of the election.
ARTICLE VII.
MANAGEMENT.
Section 1: The affairs of the Society shall be managed by a
Council under this Constitution and under the By-Laws adopted
for the execution thereof. The Council shall direct the business
of the Society either itself or through its officers and com-
mittees.
Section 2: The Council shall consist of the officers of the
Society and of the two junior Past-presidents.
(a) Sec. 2: Regular meetings of the Council shall be held
once each month, except during July, August and September.
Special meetings of the Council or of the Executive Committee
may be called by the President. Notice of such special meetings
shall be forwarded to the members of the Council or of the
Executive Committee at least three days in advance of the
meeting. The notice shall contain a synopsis of the business
to be brought before the special meeting, and no business other
than that so specified shall be transacted at such meeting.
(b) Sec. 2: The General Secretary shall, after each meeting
of the Council, forward to each member thereof a transcript of
the minutes of the meeting.
Section 3 : The Council may delegate any or all of its powers
to an Executive Committee of five members, consisting of the
President, the Secretary and the Treasurer, ex-officio, and two
other members of the Council, which committee shall conduct the
affairs of the Council between its meetings.
(c) Sec. 3 : Should the Executive Committee have taken any
action between meetings of the Council, it shall report such
action at the first meeting of the Council 'following; if approved,
the action of the Executive Committee shall be as if the action
of the Council.
By-laws are printed in small type.
16 TRANSACTIONS I. E. S. — PART I
Section 4: The President shall have general supervision of
the affairs of the Society under the direction of the Council. He
shall preside at the meetings of the Council at which he may be
present and shall be ex-officio member of all committees. He
shall deliver an address at the annual convention of the Society.
Section 5: Vice-presidents or Directors, in order of seniority,
shall preside at meetings of the Council in the absence of the
President.
Section 6: The Treasurer shall be the custodian of all
moneys. He shall make an annual report, which shall be au-
dited, and such other reports as may be prescribed. The Treas-
urer and the Secretary, with the advice and consent of the Com-
mittee on Finance, shall invest such funds as may be ordered by
the Council. They shall pay all bills when audited by the Com-
mittee on Finance and approved by the Council.
Section 7: The General Secretary shall be, under the direc-
tion of the President and the Council, the executive officer of the
Society. He shall prepare the business for the Council and re-
cord the proceedings thereof. He shall collect all moneys due
to the Society, and deposit the same subject to the order of the
Treasurer. He shall personally certify the accuracy of bills or
vouchers upon which money is to be paid and shall draw and
countersign all checks, which shall be signed by the Treasurer
when such drafts are known by him to be proper, duly authorized
by the Committee on Finance and in accordance with the neces-
sary vouchers transmitted by the General Secretary with the
draft. He shall have charge of the books and accounts of the
Society and shall furnish monthly to the Council a state-
ment of receipts and expenditures and monthly balances. He
shall present annually a report to the Council for publication in
the Transactions, and from time to time shall furnish such
statements as may be required. He shall conduct the corre-
spondence of the Society and keep full records and perform such
other duties as may be assigned to him. The Council may ap-
point assistants to the General Secretary ; one of these may have
the title of Assistant Secretary, and shall be under the immediate
direction of the General Secretary and aid him in all matters.
>?
TRANSACTIONS I. E. S. — PART I
In the event of prolonged absence or disability of the General
Secretary or Treasurer the Council shall authorize one of its
members to sign or countersign checks.
(d ) Sec. 7 : The accounts of the General Secretary and the
Treasurer shall be audited annually just prior to the annual
meeting.
Section 8: The President shall, at the first meeting of the
Council after he assumes office, appoint, subject to the approval
of the Council, the following standing committees : a Commit-
tee on Finance, of three members; a Committee on Papers, of at
least five members ; a Committee on Editing and Publication, of
three members. He may also appoint temporary committees
from time to time. Two of the three members of the Finance
Committee shall be members of the Council, and the other stand-
ing committees shall include at least one member of the Council.
(e) Sec. 8: The Council shall appoint a General Board of
Examiners to pass upon applications for membership received
from persons not residing within the territory of any section.
Section 9: All committees shall be directly responsible to the
Council, and shall act under its direction. The Council may at
any time, at its own discretion, remove any or all members of a
committee, and thereupon the President shall forthwith appoint
others as hereinbefore provided ; in the failure of the President
duly to appoint such a committee, the Council may make the ap-
pointment. The terms of the members of all standing and tem-
porary committees shall terminate at the time of the first Council
meeting of the new administration of each year. In case of fail-
ure to appoint new standing committees on Finance, on Papers
and on Editing and Publication, the retiring committees shall
continue to act until their successors are appointed.
(f) Sec. g: So far as possible, all reports of committees to
the Council shall be in writing and signed by all the members
of the Committee, or an explanation shall be offered by the
chairman for the absence of any signature. If only an oral
report of committee work can be rendered, the chairman or
other member making such report shall state if the subject mat-
ter has been submitted to the other members of the committee.
and shall offer an explanation if this has not been done.
By-laws are printed in small type.
18 TRANSACTIONS I. E. S. — PART I
Section 10: The Committee on Finance shall have direct su-
pervision of the financial affairs of the Society, and shall pre-
sent to the Council an annual report on its financial condition.
It shall approve all bills before payment, and shall make recom-
mendations to the Council as to the investment of moneys and
upon all specific appropriations. No payments other than routine
office expenses shall be made by the General Secretary or Treas-
urer, except upon the authorization of the Committee on Finance.
Section 11: The Committee on Papers shall have general su-
pervision of all papers to be presented before the Society, and
shall have the duty of preparing the programs of general meet-
ings of the Society and procuring papers for presentation before
such meeting. No paper, discussion, communication or report
shall be printed in the Transactions of the Society or elsewhere
until approved by the committee.
(g) Sec. ii : The Committee on Papers may direct the Com-
mittee on Editing and Publication to make such revision as may
be considered necessary or desirable, of papers and communica-
tions offered for publication ; in case of such revision the manu-
script shall be returned to the author to obtain his consent thereto,
and should such consent be refused, the paper or communication
shall not be accepted for presentation before the Society.
(h) Sec. ii : The acceptance of a paper or communication for
presentation before the Society or any section thereof shall not
be considered a guarantee of its publication in the Transactions.
Section 12: The Committee on Editing and Publication shall
edit all discussions of papers presented before the Society or any
section thereof, and shall decide all questions of detail regarding
the publication of papers, discussions and communications. The
Transactions and other publications of the Society shall be in,
direct charge of this committee.
(i) Sec. 12: The Committee on Editing and Publication may,
at its discretion, abridge discussions for printing. The Com-
mittee shall cancel remarks that do not bear directly on the
subject under discussion, or deal in personalities or have mani-
festly a purely commercial object.
(j) Sec. 12: All papers, discussions and other matter in-
tended for publication in the Transactions shall, so far as pos-
sible, be revised and edited in manuscript and not in proof.
By-laws are printed in small type.
TRANSACTIONS I. E. S. — PART I 19
(k) Sec. 12: A revised report of any member's discussion
on any paper must be received at the general office of the Society
within ten days after it has been mailed to the member, otherwise
revision shall be made by the Editing Committee.
(1) Sec. 12: The Transactions of the Society shall be issued
monthly, except during the three summer months.
Section 13: Five members shall constitute a quorum of the
Council. The "Vote of the Council" shall be a vote of the
majority of the members present and forming a quorum, except
where a letter-ballot is prescribed, when the "Vote of the Coun-
cil" shall be a vote of the majority of the entire membership of
the Council.
ARTICLE VIII.
MEETINGS.
Section 1: The annual meeting of the Society shall be held
on the second Friday of June of each year at a place designated
by the Council, when a report of the proceedings of the Society
for the past fiscal year shall be presented by the Council, which
report shall be verified by a majority of the Council, including
the President, Treasurer and General Secretary.
Section 2: An annual convention of the Society shall be held
on a date and at a place fixed by the Council, for the presentation
and discussion of professional papers and subjects. The Presi-
dent shall deliver a presidential address at this meeting.
Section 3: Other meetings of the Society as a body may be
held at such time and place as the Council shall direct, at which
no business affecting the organization or policy of the Society
shall be transacted. Notice of all such meetings shall be sent
by mail or otherwise to all members at least ten days in advance
of a meeting.
ARTICLE IX.
SECTIONS.
Section 1: Sections of the Society may be authorized in any
State or locality where the membership exceeds 50.
(a) Sec. I : Upon petition for the authorization of a section
of the Society, the Council may accord' such authorization if the
necessary membership exists within the locality specified in the
petition.
By-laws are printed in small type.
20 TRANSACTIONS I. E. S. — PART I
(b) Sec. i : Meetings of sections shall be held at times and
places fixed by the Board of Managers. When suitable papers
or lectures are available, meetings may be held preferably monthly
except during the three summer months.
(c) Sec. i : The meetings of the sections shall be held prefer-
ably before the 15th of the month.
Section 2: Each section shall nominate and elect a Chair-
man, five Managers and a Secretary.
Section 3: The officers of a section shall be elected annually
by the members affiliated with the section, the election to be in
accordance with a procedure fixed by the Council.
(d) Sec. 3: Procedure in nominating and electing section offi-
cers shall be as follows, except when other procedure shall be
authorized by the Council :
A section nominating committee shall be appointed by the
Section Board of Managers at a meeting held not later than
March 1 of each year. The appointment shall be reported to
the General Secretary. This committee shall consist of five
members of whom at least two shall be past officers of the
section or members of the Council.
Not later than March 15 of each year, the General Secretary
shall notify the chairman of the committee that it is the com-
mittee's duty to prepare a nomination ticket containing the
names of those whom they deem best suited for the section
offices to be filled at the ensuing annual election. The report
of the committee shall be prepared in duplicate, one copy shall
be submitted to the chairman of the section and the other copy
shall be delivered to the General Secretary not later than April
25. The ticket thus prepared by the committee on nomination
shall be printed and forwarded to all section members not later
than May 5, in connection with the ballots for election of general
officers. The election of section officers in other respects shall
be carried out in a manner similar to that prescribed for the
election of general officers, save that a copy of the report of the
Committee of Tellers on the results of the section election
shall be mailed as soon as prepared, to the chairman of the
section and to the Chairman-elect.
Section 4: The business of a section shall be conducted by a
Board of Managers, which shall consist of the Vice-president of
the Society representing the locality of the section, and the Chair-
man, Managers and Secretary of the section.
By-laws are printed in small type.
TRANSACTIONS I. E. S. — PART I 21
Section 5: The Section Board of Managers shall annually,
at the first meeting of the society year, appoint a Board of Ex-
aminers to pass upon applications for membership.
(e) Sec. 5: The Board of Examiners of a section shall con-
sist of the Chairman, the Secretary and one Manager of the
section.
Section 6: A section may formulate by-laws for its con-
duct, which shall conform with the Constitution and By-Laws
of the Society and with the policy of the Society as fixed by the
Council. Upon approval by the Council, proposed By-Laws
may be adopted by a two-thirds vote at a regular or special meet-
ing of the section ; notification of such meeting, together with a
copy of the proposed by-laws shall be sent to all members of the
section at least ten days prior to the date fixed for its holding.
Section 7: Any proposed action of a section not relating to
the holding of meetings and the discussion of papers shall be
submitted to the Council of the Society for approval prior to
being put into execution.
Section 8: The expenses of sections incurred for postal-card
notices of meetings shall be paid from the general fund of the
Society. In cases where there is no desirable auditorium avail-
able free of charge, the Council shall authorize the rental of a
hall, the expense to be payable from the general fund of the So-
ciety. Other expenses than these to be payable from the gen-
eral fund of the Society must first be authorized by the Council
of the Society.
(f) Sec. 8: The Treasurer may deposit with the Secretaries
of sections a sum of money, the amount to be fixed by the
Council, to provide for current expenses.
(g) Sec. 8: The General Secretary of the Society shall supply
to each section all stationery and printing, aside from postal-card
notices necessary for the conduct of its business.
Section 9: A Section Board of Managers may authorize, and
shall provide for the payment by local assessment of any ex-
penses of a section beyond those authorized to be paid from the
general fund of the Society.
By-laws are printed in small type.
22 TRANSACTIONS I. E. S. — PART I
Section 10: Papers shall be approved by the Section Board
of Managers prior to presentation before a section. Manu-
script of papers approved should be forwarded to the
Committee on Papers sufficiently in advance of date of presen-
tation to enable advance copies, if a paper be approved by that
committee for general presentation, to be printed and sent to all
sections for distribution prior to presentation before the sections.
Section 11: Reports of discussions shall be forwarded
promptly to the General Secretary who shall mail them at once
to members for revision.
(h) Sec. II: The Secretaries of sections shall, after each
meeting, send to the General Secretary a statement of the at-
tendance and of the business transacted.
(i) Sec. ii : The Secretary of each sction shall forward to
the General Secretary, not later than five days after a meeting of
a section, the proceedings of the meeting for publication in the
Transactions.
(j) Sec. ii : The Secretaries of sections shall send monthly
to the General Secretary an account of all expenditures in the
preceding month.
Section 12: Should the membership of a section fall below 50,
or the average attendance at meetings not warrant the expense
of maintaining the organization, the Council may cancel its
authorization.
Section 13: Sections shall abide by the Constitution and By-
Laws of the Society and conform to the regulations of the Coun-
cil. The conduct of sections shall always be in conformity with
the general policy of the Society as fixed by the Council.
ARTICLE X.
LOCAIv REPRESENTATIVES.
Section 1: When authorized by the Council, the President
shall appoint, subject to the approval of the Council, local secre-
taries or local committees resident in cities or localities where it
may be deemed desirable to provide representation with a view
to promoting the work of the Society.
By-laws are printed in small type.
TRANSACTIONS I. E. S. — PART I 23
(a) Sec. i : Local secretaries shall communicate to the Gen-
eral Secretary, information concerning local developments in
which the Society may be concerned; shall endeavor to promote
occasional meetings under the joint auspices of the Illuminating
Engineering Society and local organizations with a view to
fostering interest in the work of the Society and shall in any
other manner which may commend itself seek to develop local
knowledge concerning the objects of the Society and to advise
the General Secretary when opportunities arise for the Society
to promote its objects.
(b) Sec. I : Local secretaries may obtain Society stationery
upon application to the General Office. Local secretaries' expenses
for the correspondence may be billed to the Society.
ARTICLE XL
GENERAL.
Section 1: The fiscal year of the Society shall be October i
to September 30.
Section 2: A quorum of the Society shall consist in number
of one-tenth of the total number of members as listed in the
Society's records at the close of the last fiscal year.
ARTICLE XII.
AMENDMENTS AND BY-LAWS.
Section 1: Proposals to amend this Constitution shall be
made in writing to the Council and signed by at least
100 members and shall reach the General Secretary not
later than April 1. The Council shall consider such
proposals and direct the General Secretary to send out
a letter-ballot on their adoption. Votes to be considered
shall be received not later than May 26, and shall be re-
ferred unopened to the Committee of Tellers who shall count
such votes and make a sealed report, which shall be presented
at the annual meeting. An affirmative vote of two-thirds of the
entire vote cast by qualified members of the Society shall be
necessary to secure the adoption of an amendment. An amend-
ment shall take effect twenty days after its adoption.
By-laws are printed in small type.
24 TRANSACTIONS I. E. S. — PART I
Section 2: By-Laws in interpretation of the spirit and letter
of this Constitution and for its execution may be adopted by a
majority vote of the entire Council. Votes on by-laws shall be
by letter ballot. Each by-law proposed or adopted shall state
the article and section of article of the Constitution to which
it relates.
(a) Sec. 2 : A proposed by-law shall not be acted upon at
the same meeting of the Council at which it is submitted. At
least ten days before the Council meeting at which a by-law
will come up for definite action, a copy of the same shall be
forwarded to each member of the Council.
By-laws are printed in small type.
TRANSACTIONS
OF THE
Illuminating
Engineering Society
JUNE, 1913
PART II
K
V
Papers, Discussions and Reports
[ JUNE, 1913 ]
CONTENTS -- PART II
Some Home Experiments in Illumination from Large Area
Light Sources. By Herbert E. Ives 229
Gaslighting in an Exhibition Hall. By Robert F. Pierce. • 263
Metal Reflectors for Industrial Lighting. By Thomas W.
Rolph 268
Vision as Influenced by the Brightness of Surroundings.
By Percy W. Cobb 292
A Practical Solution of the Problem of Heterochromatic
Photometry. By Ch. Fabry 302
*>
°l
SOME HOME EXPERIMENTS IN ILLUMINATION
FROM LARGE AREA LIGHT SOURCES.
BY HERBERT E. IVES.
Synopsis: The experiments here described were carried on with the
consideration of efficiency subordinated. The problem was to obtain a
pleasing satisfactory illumination, regardless of cost or of established
ideas. Chief attention was paid to direction, diffusion and absence of
glare. Means of measuring these qualities not being established, the
criterion adopted was the writer's judgment. An analysis of light
sources is made on the basis of size, intrinsic brilliancy, direction of light
and whether the principal light source is visible or concealed. A study
of a case of satisfactory daylighting from windows shows the window
to be essentially a large area concealed light source at the side. A number
of experimental artificial windows are described, leading to an estimate
of the cost of an exact copy of daylight.
INTRODUCTION.
There is an old saying that "shoemakers' children go barefoot,"
and it is a matter of common observation that illuminating en-
gineers have little time to study the lighting of their own homes.
This should not be, of course, for a "doctor should have faith
in his own medicine." But, over and above such considerations,
is the fact that the least cultivated, but probably most important,
field for good work in illumination is the lighting of the home.
It is an almost everyday experience for us to see houses which
are beautiful by day, but at night are actual atrocities — monu-
ments to a lack of the most elementary conception of the lighting
problem on the part of the architect.
It was in the effort to light my own home satisfactorily that the
experiments to be described were begun. As time went on, how-
ever, the problems studied became somewhat varied, until lately
attempts to copy daylight distributions of illumination have occu-
pied most of his attention. The matter which follows is some-
what rambling, and the experiments somewhat incomplete, for
reasons given ; but as the next stage of the work may be de-
layed for some time it was thought well to publish this much, if
for no other reason than for the discussion which may result.
* A paper presented at a meeting of the Philadelphia section of the Illuminating
Engineering Society, May 16, 1913.
23O TRANSACTIONS I. E. S. — PART II
Out of this study have come some novel lighting devices and
some new lighting schemes which are thought worth recording.
Also, in the course of the work brightness measurements of dif-
ferent conditions were made which bring out points of interest.
ADVANTAGES AND DISADVANTAGES OF
HOME EXPERIMENTS.
A great advantage of using one's own house as an illumina-
tion laboratory is that one can live with the lighting system and
learn how it wears, as the various normal activities are carried
on at their normal times. The home rooms have their furniture
in place, which is a tremendous advantage. Everyone who has
moved to a new house knows how very different a room appears
before and after furnishing. It is the difference between dis-
comfort and comfort, and to attempt to judge a lighting scheme
in a bare laboratory room means too serious a handicap. Often
a mere touch of decorative treatment will reclaim a lighting
device which, in its crude form, is hard to imagine desirable.
The disadvantages are several. A considerable one is that it
is difficult and often impossible to make changes in the positions
of the outlets. A certain amount of flexibility is achieved by
using base-board connections, but after a more or less unsightly
structure of wires or tubes has arisen, the time is likely to come
when something logically next on the program cannot be tried
out without rebuilding the house. This limits experimentation,
consequently some of the things described in this paper are in-
complete, and the descriptions must end with suggestions as to
what might be better. The usual separation from proper tools
and the time taken to make what should be trivial changes consti-
tute other disadvantages. A special house with furniture com-
plete would possibly be the right thing for a study of this kind, but
even this would labor under the disadvantage that one would be
likely not to feel "at home," and would not be apt to live in it
long enough to determine how it wore.
A CHARACTERISTIC OF HOME LIGHTING.
A characteristic of home lighting which makes it well worth
studying by a lighting company is that efficiency counts for little.
Furniture and pictures are not bought for their cheapness,
IVES: SOME HOME EXPERIMENTS IN ILLUMINATION 23I
neither should light be — nor will it when the public becomes bet-
ter educated in the use of light and in its remarkable decorative
possibilities. In domestic lighting the real problem is to obtain
a pleasant illumination effect, almost without regard to cost. In
the experiments here described no attention whatever has been
paid to efficiency. The author has put himself in the position of
a householder who demands certain lighting effects and is willing
to pay for satisfaction. In fact, I believe that many possible
lighting schemes have not been tried at all — notably copies of
daylight distribution — because the illuminating engineer has been
apt to think too quickly of efficiency. He has been an engineer
before being an artist, whereas the more fruitful procedure is to
first obtain the desired effect and then count the cost. It may
happen that the price of the paint prevents the painting of a mas-
terpiece. Even if it costs too much now to have just what we
would like, we must look forward to the day of more efficient
light sources. We may even bring that day nearer by finding
greater needs for them.
MEASUREMENTS AND MEASURING INSTRUMENTS.
This paper is to be comparatively free from foot-candle values,
watts per square foot and the like. The chief points aimed at
are diffusion, proper direction, freedom from excessive contrast
and glare, a pleasant non-fatiguing quality and good appearance.
How can these be measured? In the writer's opinion the only re-
liable and sensitive instrument is the experienced eye of one who
has observed and thought about illumination for a period of
years. We have all of us noticed how sensitive we have become
to exposed light sources. Formerly they irritated us, to be sure,
but we did not know the cause. Now we choose our position with
care, shield our eyes with our hands, and in other ways give
testimony to the vastly increased sensitiveness of our seeing
mechanism. Meanwhile it remains a sad fact that we have gen-
erally available no method of measuring lighting conditions
which will do more than distinguish between two so extreme in
quality that a mere casual glance will tell which is good and
which is bad. Until we are better off in this respect the trained
eye must be encouraged as the best we have. It is worse than
3
232 TRANSACTIONS I. E. S. — PART II
useless to make measurements of factors which are not the ones
really vital. If home illumination demands diffusion and a feel-
ing of comfort meter-candles illumination on the working plane
has very little to do with the problem. It must be borne in mind
that measurements are of no value unless they record conditions
whose qualities are proved by experience. Experience can never
be dispensed with. It must be called upon sooner or later. This
paper deals with lighting schemes whose measuring instrument
at present is experience. Recognizing this fact I have out-
spokenly given here as the final criticism my own judgment of
the success of an installation. In all cases however, I have had
as many comments and suggestions from others as I could obtain.
REMARKS ON DIRECTION, DIFFUSION, GLARE
AND CONTRAST.
One peculiarity of daylight illumination from windows is its
direction. Most artificial systems cast the light downward from
a point near the center of the ceiling. There then occurs in
changing from day to artificial light a 90 degree rotation of shad-
ows. By day they are long and sweep across the room ; by night
they are short or completely covered by the shadow-casting
object. A clear illustration of this change is to be seen in the
ordinary railway car. By day the illumination comes from the
windows at the two sides ; at night from the high centrally placed
units. As far as direction is concerned, side wall brackets, floor
standards or table lamps on side tables approximate more nearly
to window conditions.
Diffusion of light with soft shadows, as is well known, is most
perfectly brought about by the use of a large area of light source
and is one of the chief merits of the "indirect" system. A cer-
tain degree of diffusion may also be obtained by a multiplicity of
light sources, although this is apt to give merely a multiplicity of
sharply defined shadows, instead of the soft shadows of the large
source. The ordinary window with its area of 5 or 6 square
feet (0.46 or 0.56 sq. m.), gives a far better degree of diffusion
than most artificial schemes. Another factor in diffusion is the
color of the light. Blue light diffuses better than yellow. It is
therefore quite possible that daylight is inherently apt to be better
diffused than is yellow artificial light.
IVES: SOME HOME EXPERIMENTS IN ILLUMINATION 233
In regard to contrast and glare, it is well to bear in mind some
numerical data. A Welsbach mantle has an intrinsic brilliancy
of 35 candle-power per square inch (6.45 sq. cm.) ; a tungsten
filament about 1,000 and a patch of sky about 2 to 3. For the
same intensity of illumination, that is, the same general bright-
ness of illuminated objects, therefore, the physical contrast be-
tween the brightness of the illuminated objects and the light
sources themselves will be from ten times to hundreds of times
greater with visible "direct" artificial light than with daylight.
If the light sources are concealed, this contrast may be avoided,
but it occurs again to a disturbing degree if specularly reflecting
surfaces are present. These reflect images of the light sources
with about 1/10 or V20 °f tne intrinsic brightness of the parent
source. It is, in fact, immediately evident to an expert whether
the lighting of a room is due to small or large sources by an
inspection of these two things — the sharpness of the shadows and
the brightness of the specular reflections. Half of this handicap
of small sources is eliminated by entirely avoiding specularly
reflecting surfaces.
SOME CHARACTERISTICS OF DAYLIGHT OUT-OF-DOORS.
Out-of-doors daylight may be pleasant or unpleasant. In a
previous paper* which records brightness measurements of
typical outside illuminations, certain conditions were found to be
most pleasing. These were an excess of brightness in the upper
hemisphere, usually with a maximum near the horizon, and (a
very essential condition) sunlight directed from the side to give
long shadows.
A criticism which has been made of many "indirect" installa-
tions is that the bright "sky" furnished by the light ceiling is
bounded by dark walls carried up above the eye line. This gives
an effect of being down in a well, or as though the room had
been lifted up away from one. To produce a real out-of-doors
daylight distribution, the above quoted work would call for an
extension of this brightness down to the horizontal. This would
tend to overcome an objection frequently made to the 'indirect"
system — that faces are subject to unnatural downward shadows.
* The Distribution of Luminosity in Nature, Trans., I. E. S., p. 6S7, Vol. VI (1911).
234
TRANSACTIONS I. E. S. — PART II
Out-of-doors our faces receive considerable horizontal illumina-
tion. In addition the habit of wearing hats must be taken into
account. Hats protect the eyes from overhead bright areas and
also reduce the vertical component of the illumination on the
face.
The effect of the large sky angle on the extreme brightness
ratio must not be overlooked. Out-of-doors the sky approxi-
mates to an infinite plane. Under this a white surface will be as
bright as the illuminant while other objects will be bright in
proportion to their reflecting powers. As the area of "sky" is
decreased, its brightness remains the same, but that of illuminated
objects becomes less, thereby increasing the extreme brightness
ratio between unconcealed illuminant and illuminated objects.
Diffusion is also decreased and shadows take on a more definite
downward direction.
AN ATTEMPT TO COPY OUT-OF-DOORS DIFFUSED DAYLIGHT.
The guiding idea here was to carry the direct illumination of
an "indirect" system down on to the walls in such manner as to
approximate the out-of-doors conditions, and thereby produce
a pleasant diffused illumination. The room available was a
Fig i. — Diffused illumination from side walls.
small one, about 14 feet (4.27 m.) square, furnished with a cen-
tral fixture and papered a medium light buff. Under this central
fixture was suspended a yellowish Japanese umbrella of oiled
paper, convex side up. The direct light of the lamps thus fell
on the walls, which acted as the chief light source of the room,
since the transmission of the umbrella as used was low, while the
IVES: SOME HOME EXPERIMENTS IN ILLUMINATION 235
ceiling was little brighter than the walls and largely invisible be-
cause of the comparatively small size of the room. (See Fig. 1.)
A peculiar merit of this scheme lay in the splendid illumination
of the pictures on the walls. These had to be slightly tilted to
prevent reflection of the bright light sources, and in the direction
of the entrance doorway a translucent Japanese fan was placed
above the umbrella to conceal the lights. This installa-
tion was used for many months and gave great satisfaction.
The "well" appearance sometimes noted in indirect systems was
entirely absent, There was no tendency for the eye to wander
upward to a bright point above. Although the illumination
seemed low, it proved ample for sewing and other close work,
which was done without the fatigue customary to working with
"artificial" light. A pleasing addition to the general diffused
lighting was furnished by placing a table lamp on one side,
whereby the direct light of the sun was copied. One point
proved to my satisfaction by this experiment is that light walls,
well illuminated, are not productive of evil. On the contrary,
they help in a marked degree the general effect of brightness. A
room looks brighter with bright walls and, unless some extreme
case is taken this means that the object — to light the room — has
been attained.
This lighting arrangement was only given up when I moved
away from Cleveland. It needs a small room, or, in a large room,
a number of ceiling outlets not far from the walls in place of one
central one. It is intended at some future time to place outlets in
a large room which will make possible lighting the walls in a
similar manner and the expectation is that it will furnish a very
satisfactory illumination. In some indirect or semi-indirect in-
stallations it might be of interest to tilt the units near the walls
in order to secure a similar effect. These units might be made
purposely unsymmetrical.
CLASSIFICATION OF LIGHT SOURCES.
The common classification into direct, indirect and semi-in-
direct, while convenient, is by no means a complete analysis. The
true classification is on the basis of the size of the light source
and whether it is concealed or visible. The earlier light sources,
236 TRANSACTIONS I. E. S. — PART II
such as the gas flame and the carbon filament were small sources
of relatively high intrinsic brilliancy. From them we have been
working toward large light sources of lower intrinsic brilliancy.
The fact that some of these are large areas of diffusely reflecting
ceiling and that others are large areas of diffusely transmitting
glassware is of no great significance. It is of significance that
the light source is of large extent and of low intrinsic brilliancy.
It is also of significance that in each of these cases ("indirect"
and "semi-direct") the principal light source is visible. By
"principal light source" I mean the source which furnishes the
greater part of the light which falls on floor, furniture and work-
ing plane. It is also of great importance that as ordinarily
worked out, in "indirect" and "semi-indirect" installation, the
light source is of necessity the brightest object in the field of
view. Therefore, the indirect and the semi-indirect systems sift
down to this : that the unconcealed light source is made as large
as is possible in the attempt to make its intrinsic brilliancy low
and bearable. Incidentally this causes great diffusion of light
and soft shadows.
In some forms of "direct" lighting another device is employed,
namely, partial or complete concealment of the light source. In
this case it may result that the principal light source is not the
brightest object in the field of view. If the concealment is suffi-
cient there may be a complete absence of points of high in-
trinsic brilliancy — more so than in the "indirect" system — except
for points of specular reflection of the light sources if polished
surfaces are present. These latter are of high intrinsic bril-
liancy in the "direct" systems and low in the "indirect."
It will be noted by those who have followed this analysis that
while we have lighting systems with large, visible, low-brightness
sources, and systems with small, either visible or invisible, high
brightness sources, there are none consisting of large, low bright-
ness concealed light sources — (with the possible exception of
some forms of deck lighting where the light is carried well above
a transparent or only partly diffusing glass). As will be pointed
out later, this is the specification (when that of direction from
the side is included) of pleasant daylight illumination from win-
dows.
IVES: SOME HOME EXPERIMENTS IN ILLUMINATION 237
A LARGE OVERHEAD LIGHT SOURCE.
It follows from the classification just given that the charac-
teristics of "indirect" lighting would be obtained without the
process of reflection if a large enough low-brilliancy source was
used. This was experimentally accomplished in a very simple
manner by the use of a yellowish oiled paper Japanese umbrella
of 4 feet diameter, turned point down, within which are the lights
in frosted globes. (See Fig. 2.) A certain amount of light
falls on the ceiling, so that it might be called a "semi-indirect"
system, but by far the greater part of the illumination is due to
the visible light source. This is of rather low intrinsic brilliancy,
so that it can be looked at without any sensation of strain, and
gives ample light all over the 16 foot by 20 foot room with an
emission of about 7,500 lumens.
Fig. 2. — Illumination from large area, low brightness overhead source.
Now as to its merits and defects. Because of the size and low
intrinsic brilliancy shadows are soft and specular reflections are
too dull to be annoying. Because of the conical shape a better
illumination of the walls and far corners is obtained than would
be possible by a ceiling reflection scheme. On the other hand,
the direction of the shadows is, for the center of the room at
least, straight down and, therefore, different from the most
agreeable daylight condition. But the greatest defect lies in the
fact that the principal light source is visible, and, low as is the
238 TRANSACTIONS I. E. S. — PART II
intrinsic brightness, it is the brightest object visible and is too
bright to be continuously in the field of vision. When the occu-
pants of the room sit around the center table or sit sideways to
the light, the illumination is extremely satisfactory. If, how-
ever, when a number of people are conversing, they face each
other across the center, then in time the large bright umbrella
becomes an irritant — far less, of course, than would the usual
bright points, but still noticeable to a sensitive and critical eye.
An experiment was tried with a still larger light source, but
it had the same defects. The conclusion was arrived at that by
no process of increasing the size of the source, unless it actually
occupied the whole upper hemisphere — which is out-of-doors
daylight on a sunless day — could the light source be made in-
nocuous. Diffusion is not alone sufficient. Any system by which
the light source is the brightest visible thing gives an extreme
brightness ratio which is too great to wear well. If, in the
room illuminated with the large umbrella, the hand is held over
the eyes, or an eye shade is worn, it leaves little to be desired ;
there remains merely the question of the direction of the light
to be objected to; indeed many might prefer the centrally located
source. Of course the higher the light source the less this de-
fect would be, so that the "indirect" practise of making the ceil-
ing the real source chooses the best condition, for it then approxi-
mates, in a small room, to a concealed source.
What seems to be most desirable in this particular room
is some means of illumination with this large low brightness
source, at the same time keeping it concealed. That done, the
range of visible brightness is not too great. As will be seen
below, daylight illumination by windows does something of this
sort.
CHARACTERISTICS OF PLEASANT DAYLIGHT ILLUMINATION
FROM WINDOWS.
I have no brief for daylight illumination in general, nor for
all kinds of window lighting. Excessively unpleasant illumina-
tion may be found by day. On the other hand, daylight at its
best has never, in my opinion, had a rival for general excel-
lence. There has been much discussion on this point, some
IVES': SOME HOME EXPERIMENTS IN ILLUMINATION 239
declaring flatfoot for daylight as necessarily best, others record-
ing themselves as preferring the "mellow," "cozy" artificial light.
As a matter of fact, I believe no one has ever had a good copy
of daylight to use at night in place of the ordinary artificial
systems. Perhaps it is a case of "sour grapes" with those who
say they would not copy daylight if they could. I believe, too,
that illuminating engineers have neglected analysis of the good
and bad points of daylight. Without such analysis it is futile to
hope to copy the merits or avoid the defects.
My observation, common to that of others, is that window
lighting is not always satisfactory. It may be insufficient, or it
may cause excessive contrast in brightness. In my own house
the first floor living room has one door and one window on the
front and two windows at the side facing another house. Ordi-
narily the room is not sufficiently lighted by daylight. The front
window is too narrow. If the shade is run up to the top it is im-
possible to look with comfort at the window, because the large
patch of bright sky then visible is in violent contrast to the unil-
luminated wall at the sides of the window. Even with the
shades down, so that the sky is not seen, the houses opposite are
apt to present too great contrast of brightness to the walls adja-
cent to the window. Only under one condition is the daylight
really satisfactory and that is when a spot of sunlight falls well
back on the floor, the houses opposite being in the shade. Then
the walls adjacent to the window receive light from the floor, the
ceiling receives light from the floor and from the pavement out-
side and all is bright and cheerful. With snow on the ground
outside, the effect is still better.
Another case of bad daylight illumination is furnished in the
upper floors of a tall office building which is not faced by any
other structure. If one sits facing a window, one sees a large
expanse of bright sky, which soon becomes painful. In these
rooms the official habitually places his desk so that his back is
toward the windows. Visitors who face the windows do not find
daylight pleasant.
Most fortunately a case of the other kind is to be found in my
own house, namely, an instance of a pleasant and satisfactory
daylight illumination. This is in a second floor room, facing
240
TRANSACTIONS I. E. S. — PART II
south, illuminated by two adjacent windows in a bay nearly the
width of the room. The wall paper is a figured gray of approxi-
mately 45 per cent, reflection coefficient. The dimensions of the
room are shown in Fig. 3. A long room of this type is perhaps
one of the hardest to illuminate satisfactorily by daylight or arti-
ficial daylight because of its depth. One with windows on two
sides would have been preferable and easier to deal with, if it
had been available.
/o
:.*>
Fig. 3. — Distribution of surface brightness in arbitrary units by daylight. Large values,
a case where sunlight falls on floor. Small values, sun not streaming in.
Observations, both qualitative and quantitative, made from
time to time in this room, revealed some interesting points. ( 1 )
At all times, raising the shades above the middle sashes made
the lighting less pleasant. (2) The room was brightest and, on
the whole, most pleasing when the sun streamed in on the floor
or wall and shone on the street below. (3) The effect was
unpleasant when the sun fell upon the (gray stone) house fronts
opposite. (4) A gray or misty day was not pleasant. (5) The
distribution of illumination or brightness was such that the face
of a person sitting with his back or side to the window received
sufficient light from the room so as not to be in marked contrast
to the background, or to have the two sides of his face unduly
different in brightness.
Two sets of measurements of brightness are plotted in
Fig. 3, one for the case of sunlight streaming in the window, the
IVES: SOME HOME EXPERIMENTS IN ILLUMINATION 24I
other when it is on the street below, but not streaming in nor on
the houses opposite. Both effects were extremely satisfactory.
An analysis of the various data has led me to the following
conclusions :
The most pleasant conditions are those when the effective light
source is of three parts : ( i ) A large area of bright sky sub-
tending the solid angle of the window, invisible from the greater
part of the room and illuminating the floor and lower part of the
room. (2) A large area of intrinsic brilliancy about one-tenth
that of the sky, visible from the room (houses opposite). (3) A
large area of intrinsic brilliancy about one-fourth to one-fifth that
of the sky, invisible from the greater part of the room, but illu-
minating its ceiling.
Add that these light sources are at the side, and remember
that the bright sky is of much lower intrinsic brilliancy than most
artificial light sources, and the complete specification of pleasant
window daylight of this room becomes something like this:
illumination from large area concealed light sources at the side,
that is, just the case we have seen to be absent from prevalent
lighting systems.
There must be added to this set of conditions the further one
that the walls adjacent to the window must be well lighted in
order to prevent excessive contrast. This is frequently accom-
plished by having windows on two sides. In the present case the
walls adjacent to the windows are perpendicular to these and
are, therefore, lighted by one set of windows.
Probably to an architect, who has made a study of window
lighting, this is all an old story. It is interesting in the light of
these observations to notice that low broad windows (frequently
in alcoves) are becoming common in the newer houses, taking the
place of the old-fashioned high narrow window, the upper por-
tion of which sometimes cannot be used without exposing to view
a dazzling patch of sky.
THE WINDOW AS A LIGHT SOURCE.
The numerical data given in Fig. 3 make possible a study of
the windows of this particular room as a light source.
As viewed from a point in the room the windows present prac-
242
TRANSACTIONS I. E. S. — PART II
tically the appearance of two adjacent squares 30 in. by 30 in.
(0.762 m. x 0762 m.), with their centers at a height of 42 in.
(1.07 m.) above the floor. As viewed from various points the
angle subtended by the light source of course changes, diminishing
to zero above and below and to either side. Were the landscape as
seen through the window of uniform intrinsic brilliancy, then the
window would be equivalent to a flat uniformly bright plate. It
,^«S*
U4j>AJ
Fig. 4.— The window considered as a point source of light.
could be copied by a sheet of thick opal glass illuminated from
outside, or by a diffusely reflecting surface, such as an "indirect"
ceiling source turned through 900. This, however, is not the
case, except when in a high building facing the sky, or if a fog
reduces all out-of-doors to a nearly uniform gray (a ratio of
sky to houses of 2 to 1 was found by measurement on a foggy
IVES: SOME HOME EXPERIMENTS IN ILLUMINATION 243
day). Both of these conditions are found to be undesirable, as
noted above.
In the second-story room, which was studied chiefly, the
change of intrinsic brilliancy and effective area may be followed
best by consulting the drawing, Fig. 4, in which the window is
considered, for convenience only, as a point source of light. Start-
ing directly below the window the illumination is from a narrow
slit of the intrinsic brilliancy of the sky. As the test surface is
moved away from the window this slit increases in width (as the
cosine of the angle with the normal) until a point (A) well back
on the floor is reached. Here the top of the houses opposite be-
comes visible, cutting off part of the sky. At the point (B) the
sky is no longer visible and from there to the point (C) high up
on the wall the intrinsic brilliancy of the window is that of the
house opposite. Between (C) and (D) the street below becomes
visible and from (D) over the ceiling to the window the illumina-
tion is from a patch of the brightness of the street surface. There
are. therefore, to be distinguished (in this case) three different
brightnesses, the distribution of illumination from each of which
is represented by a circle of appropriate diameter, tangent to the
window. From the brightness measurement of Fig. 3, it ap-
pears that these three circles should have diameters for the most
pleasant condition of about 10-1-3, and they are so represented
in the figures.
It will be seen that a seated person in almost any part of the
room, looking toward the window, sees only the comparatively
dull houses opposite. The floor and ceiling are illuminated by a
much brighter source than the houses and act as secondary light
sources of low brightness. If the sun strikes the floor the spot
becomes another secondary light source. The most startling
thing is the very irregular shape of the light distribution curve
and its sharp transition from maximum to minimum, unequalled,
I believe, in any commercial lighting unit. Were it not for the
fact that the source (window) is large these sharp transitions
would cause sharp contrasts of light and shadow between the
portion of wall and ceiling represented at A, B, C, etc. Actually,
the deviation of the window from a point source makes the
transitions gradual and soft.
244 TRANSACTIONS I. E. S. — PART II
An interesting feature of the distribution from a window is
that it tends to diminish the extreme brightness ratio. The great-
est illumination is on the floor and furniture, which are almost
invariably the darkest surfaces. With a uniform distribution
from a side window or with many artificial systems the floor
and furniture retain the normal brightnesses which their low
reflecting powers give them. With windows they are made sev-
eral fold brighter, which must be no inconsiderable factor in the
general effect. This fact explains in part why with "indirect"
systems the floor and furniture are apt to appear unnaturally
dark. In order to preserve the daylight ratio of brightness above
and below it would be necessary with the "indirect" system, with
its bright ceiling, to exchange the floor coverings and furniture
for some with at least a 4 or 5 fold higher coefficient of reflec-
tion. The excellent appearance of white tiled barber shops un-
der the "indirect" system is confirmatory of this.
ATTEMPTS TO ARTIFICIALLY APPROXIMATE THE CHOSEN
DAYLIGHT DISTRIBUTION.
The General Problem. — As an interesting exercise, attention
was turned to ways and means for reproducing artificially the
best daylight conditions in the room described. It soon became
evident that the problem was not a simple one. It appears, in
fact, that there is just one solution, and that is to reproduce en-
tire the broad expanse of sky, houses and street. No small light
source will copy all the qualities desired. For instance, while it
is possible to produce at a given point in the room the general
distribution of illumination and brightness given by daylight, the
same lighting scheme will not produce the desired effect at an-
other point. The problem is closely akin to that of producing
a perspective drawing which would automatically change its per-
spective as the observer moved, or a photograph in which one
could look around the corners by moving one's head from side to
side. For instance, a horizontal plate placed above and outside
the window would produce the desired sky effect near the win-
dow, but would fail farther back in the room. A vertical plate
of low intrinsic brilliancy would take the place of the opposite
houses as far as the eye of an observer within the room, but the
1VKS: SOME HOME EXPERIMENTS IN ILLUMINATION" 245
sweep of tenfold brighter light across the floor and the brighter
light on the ceiling would be missing. As it was not possible to
light up "all out of doors," several approximations to the condi-
tions given in Fig. 3 were attempted, which are described below.
The first of these may be called a lighting fixture, the others
lighting schemes, or, as a non-technical visitor described one of
them : "not lights, but light."
(i) A Side Wall Fixture to Approximate a Window Effect. —
This fixture might be popularly described as "semi-indirect from
the side," although it presents a number of deviations from what
would be obtained by merely supporting a semi-indirect bowl out
from the wall. In construction it might be most nearly described
by saying it is a table lamp with the back half and top of the
shade removed. In detail it consists of the light source, which
should be of such form or provided with such a reflector as to
throw its light chiefly below and above the horizontal about as
does the prototype window curve, added to which is a large area
translucent screen so calculated as to let the direct light fall upon
the wall behind, the ceiling above and the whole floor of the room,
and of such degree of translucency as to closely approximate the
brightness of the illuminated wall behind. The device is shown
in Fig. 5, where it is represented as a floor standard. From the
plan and elevation it will be seen that for no normal position in
the room is the bright part of the light source visible, but screen
and wall form a large source of low intrinsic brilliancy. This is
secured by the peculiar shape of the screen, which is concave to
the light with respect to the horizontal plane, half enclosing it,
but only large enough vertically to conceal the light from the
occupants of the room. The screens so far made up have been
covered with flowered silk, or cretonne, and lend themselves ad-
mirably to the general decorative scheme.
This fixture when used in a room having light walls (a
necessary condition) gives to a large degree the desired window
effect. The light is directed from the side, giving the long pleas-
ing shadows; there is a large measure of diffusion, due to the
large area of the effective source ; the floor and furniture have a
pleasant "sun lit" appearance. The thing of beauty is the light
itself and not. as in many cases, a piece of decorative metal work.
246
TRANSACTIONS I. K. S. — PART II
designed to be viewed by daylight. The limitations to this device
are the necessity for bright walls, which is not serious, and the
fact that the direct light in the lower hemisphere cannot be made
to fall as far out on the floor as with windows, without carrying
the shade too far in front of the light for convenience. It does,
' /
Fig. 5. — A side wall unit to approximate window conditions.
however, make an extremely pleasing light source and has, in
several months of use, worn well. It is especially good if two or
more are grouped at the window end of the room. It appears
that the best height for these shades, if they are to be in the field
of vision, is about on the level of the eyes. If a light source
must be visible it seems to be less irritating at this line than at
IVES: SOME HOME EXPERIMENTS IN ILLUMINATION 247
some degrees above. It is perfectly possible to make the shades
and wall behind of such brightness that a person's face may be
viewed against them as a background without undue contrast.
Side wall brackets have always been popular with those who
wish their faces to appear at their best. The defect of most
bracket fixtures has been that the light sources have an altogether
too great intrinsic brilliancy and make excessive contrast with
the wall behind, especially if the latter is dark wood, as it too
often is. This defect is overcome in the arrangement here de-
scribed. There has thus far been no dissenting opinion as to the
excellent performance of these side wall "windows."
0+
Fig. 6.— A point source copy of daylighting. (Brightness values are
multiplied by a factor 10).
(2-) An Alcove Light with Semi-window Characteristics. —
This scheme was tried out after a study of the window consid-
ered as a point source as shown in Fig. 4, and was carried
through in full knowledge of the fact that the characteristics of
a large window could not be completely copied without a com-
paratively large source. The plan is shown in Fig. 6. where
"A" represents a line of light sources with aluminized scoop
reflectors sending a large part of their light in the lower hemi-
sphere and approximating the lower portion of the large circle of
Fig. 4. At "B" is placed a frame 15 in. x 6l/2 ft. covered with
cretonne and muslin sheeting. This to an observer back in the
room takes the place of the houses opposite, as a large area low
brightness source. By trial such a thickness of fabric was found
4
248 TRANSACTIONS I. E. S. — PART II
that the illumination back in the room due to it was one-tenth
that from the row of lights as seen below it from the floor, that
is, the condition found best by daylight.
This installation is exceedingly interesting. With 7,200
lumens the surface brightnesses through the room measure up on
an average one-twelfth that found as the average of the two
daylight distributions given in Fig. 3. It was found most desir-
able, however, to let considerable light go to the ceiling, which
gives more light in the room and approximates more to the "sun-
light" case of daylight distribution in Fig. 3. This, when the eye
is adapted to night conditions, is very nearly ample at every point
of the room. Probably twice as much would be more than
enough. The room presents a strikingly daylight appearance,
except for three things, all due to the same cause. First, the line
of demarcation between the light from the 'houses opposite" and
the "sky" is too sharply marked on the walls ; second, the specu-
lar reflections on the furniture, books, etc., near the windows,
visible on entering the room, are relatively far brighter than they
would be by day, as is to be expected ; third, the shadows of the
furniture are too sharp. These are all due to the point source
character of the light, and are only to be seen above and below
the parts illuminated by the cloth covered screen. The appear-
ance is practically the same as can be obtained by covering all
the windows except a narrow horizontal slit. Wtihin the angle
which receives light only from the screen and from walls and
ceilings, the daylight character is almost startling.*
(3) A Window Reflector. — The requirement of a large-area
concealed light source at the side, having the distribution curve
of the window as a light source, may be approximated to by
combinations of optical devices, such as prisms and reflecting
surfaces. One of these devices forms the last experiment to be
described.
In accordance with the subject matter of this paper, it is clear
that an "indirect" fixture, if directed toward the wall, would not
* If the windows be made horizontal slits ol about 10 inches width and if the time of
day is taken when the sun falls on the houses opposite, an extreme case of bad lighting
is obtained. The illumination of the room is quite insufficient; there is a high ratio of
brightness between the light source and its surrounding in the field of view; shadows are
sharp and specular reflections are too bright. This illustrates clearly the importance of
the light source being large and of having its brightest part concealed.
IVES: SOME HOME EXPERIMENTS IN ILLUMINATION 249
give the desired window effect because, while the light source
would be large, it would be of uniform brightness as viewed from
all directions. In order to secure for this experiment the proper
unsymmetrical distribution of brightness, recourse was had to re-
flecting surfaces intermediate between specular and matt. Meas-
urement of a number of surfaces, such as scratch-brushed alumi-
num, aluminum paint, mirror glass covered with transparent cur-
tain material, etc., led to the choice of ribbed mirror glass, sand-
blasted on the front. Later it was found better to decrease the
brightness of the diffuse component, and this was done by rub-
bing the sandblasted surface slightly with oil. Probably a lighter
sandblasting would have given the desired quality. Such mir-
Fig. 7. — Section of special reflectors used for artificial window.
rors placed with the ribs vertical and illuminated by a light in
front and above give a large semi-specularly reflected component,
which is nevertheless so much diffused as to appear to come from
a large area of mirror; and a small diffuse component, the ratio
between the normal and reflected brightness being easily made
one-tenth by tilting the mirror.
A pair of frames of these was made, as shown in Fig. 7, each
frame consisting of four 7 in. by 30 in. elements. These frames
were hung in the window, as shown in Fig. 8, and were illumi-
nated by a row of small units in the aluminized concentrating
reflectors used before. As shown in Fig. 7, the specular reflection
from the three upper elements illuminates the floors and lower
250
TRANSACTIONS I. E. S. — PART II
parts of the walls of the room. The bottom elements illuminate
the ceiling, taking the place of the brightly lit street pavement
before.
An observer sitting in the room sees these surfaces as of a uni-
form low brightness. On dropping his head he begins to see the
top elements brighten, corresponding to the sky appearing over
the house tops. On dropping still lower, the other elements
brighten until at the floor the whole "window" appears a uniform
sheet of light. On standing up and approaching the window the
lowest elements become bright in the same manner. This copy
of window conditions is in fact very close. The reflected light
is, because of the ribs and the sandblasting, excellently diffused.
Shadows in the room are long and soft.
!
Fig. 8. — Artificial window lighting. (Brightness values multiplied by factor 10).
The excellence of the copy is shown by the relative brightness
measurements. These indicate that the relative brightness of all
visible objects is nearly the same as by daylight without the sun
on the floor, as shown in Fig 3, the actual brightness being close
to one-tenth of that of daylight. In fact this imitation of day-
light distribution in the room is, from the standpoint of relative
brightness, direction of light and diffusion, so perfect as to leave
outstanding only the factors of actual intensity of illumination
and color of light.
Good and Bad Features. — This reflector scheme has many ele-
ments of merit. It is in some respects almost uncanny, espe-
cially to one not looking toward the imitation windows. The
sweep of light over the floor; the illumination of the low por-
IVES: SOME HOME EXPERIMENTS IN ILLUMINATION 251
tion of the walls; the brightness ratio between ceiling and floor
are most striking and very different from the usual artificial light
conditions. But the general effect is not pleasing. There is a
harsh "contrasty" effect. The darker end of the room looks too
dark ; the windows top light ; shadows are too inky black.
The effect is very much like a foggy day, or as though the houses
opposite were brilliantly sunlit. Why is this? In some part to
mental bias, perhaps; the windows do not look just like real win-
dows and one is strongly conscious of the fact. Also the fact
that the light source is near, instead of distant, and that our ocu-
lar muscles of accommodation are quite conscious of the real
state of affairs. The explanation of the greater part probably
lies in two facts : first, the absolute intensity is too low, and, sec-
ond, the color of the light is not white. The absolute intensity is
about one-tenth daylight intensity. My observation leads me to
believe that a given physical ratio of brightness becomes sub-
jectively greatly increased when the order of illumination is de-
creased as much as it is here. If a ratio of brightness of 10-1
for different points in the room is satisfactory by daylight then
with 1 -10 that illumination this ratio must be decreased to 5-1
or 3-1. With this low order of illumination the shadows, such
as those on the room side of the face, are so dark as to appear
nearly black; the high-light side alone is seen. By the greater
light of day, although the relative brightness of the two sides of
the face is the same, the dark side is bright enough to be easily
seen ; the bright side does not appear too bright. I can best de-
scribe the phenomenon by saying that the room under this 1-10
illumination looks like an under-exposed photograph — the high
lights alone come out. By exposing longer a good negative is
obtained; by increasing the illumination tenfold, I believe — un-
less the color difference has a great deal to say — that this copy of
daylight would be "the real thing." Physically it measures up
in the right proportion. Subjectively the proportion becomes
distorted. I have found support for this idea by the experiment
of wearing dark glasses of 1-10 transmission in this room when
the daylight conditions are good. The conditions become trying
at once. What is called for in the present case is an increase in
total light, which cannot be made in the existing: house installa-
252 TRANSACTIONS I. E. S. — PART II
tion without danger. Incidentally it may be remarked that this
accentuation of contrasts merely means a larger value for the
Fechner fraction at low illuminations, which is actually found by
experiment.
The general effect in the room is enormously improved by add-
ing the light from two side wall table lamps described above,
placed in the corners next the alcove. These give somewhat
more light, but more than that, they throw their added light upon
the ceiling and side walls without increasing the brightness of the
"windows." These latter cease to be too bright ; the shaded side
of the face becomes visible and a first rate effect is produced.
On turning off the window light alone the room seems to jump
upward, showing very prettily the importance of the daylight
effect of the sweep of light on the lower part of the room. I be-
lieve this window experiment would have been more successful
had the room been provided with an artificial window at the side.
A long dark room such as this is difficult to light by day ; were it
not on the south side of the house I believe it might be a failure.
It certainly represents an extreme type. I am inclined to think
another type of room say a broad shallow one furnished with
these artificial windows might be quite successful, even without
an increase in the light.
The color of the light may have something to do with the harsh
effect. As above noted blue light scatters much better than yel-
low. A photograph by ultra-violet light will show almost no
shadows owing to scattered light from the dust particles in the
air, while in infra-red light shadows are black. Something of
this sort may contribute to the inadequacy of this artificial
window.
SUMMARY OF CONCLUSIONS AND SPECULATIONS ON ROOM
LIGHTING BY LARGE AREA SOURCES.
Several conclusions which seem to be justified by these various
experiments, are given as follows :
The desirable qualities sometimes found in window lighting by
daylight are, (1) direction from the side; (2) soft shadows and
low intensity of specular reflection, due to the large size of the
source; (3) the direction of a large part of the light on the lower,
IVES: SOME HOME EXPERIMENTS IN ILLUMINATION 253
usually the darker, parts of the room, and (4) the concealment of
the principal light source.
It appears possible to make a very close physical copy of the
window as a light source whereby all the characteristics may be
duplicated. An attempt to do this shows that to the above quali-
ties must be added (5) large quantity of light as compared with
what is usually available from artificial sources. It is suggested
that this is less necessary if the artificial windows are placed so
as to secure a more uniform distribution of illumination than is
necessary by daylight.
These various qualities can be separated to some extent into
necessities and luxuries; some are necessary for comfort, others
appeal to the esthetic sense. I believe that in concealment of the
light source, so that it is not the brightest object visible, in making
it of large area and in introducing a certain amount of side light
from large area, very low intrinsic brilliancy sources, most of the
really necessary and many of the attractive characteristics of
daylight may be obtained, with much lower intensity and, conse-
quently, lower cost than a true copy of daylight would come to.
I favor a combination of deck lighting for general floor and
working plane illumination, with side wall lights of the type de-
scribed above. I hope to report on an installation of this
type at some future date. If the room is not too large nothing at
present promises to excel the side wall window brackets of Fig. 5.
THE COST OF COPYING DAYLIGHT.
If it were not for the prohibitive cost of an exact copy of good
window daylight, it would be a very desirable thing. This paper
may be concluded with a calculation of just what this cost is.
The room experimented with has an area of 200 square feet,
1,600 lumens were used and ten times that were required, or
16,000. Suppose absorbing screens were used to make this light
of daylight color ; this would call for about ten times the amount
of light, or 160,000 lumens generated. Taking the tungsten lamp
at 1.5 watts per spherical candle as the light source, one watt per
square foot is actually used, ten watts is' called for, one hundred
if subtractive daylight is made (or 20 kilowatts for the room).
Let us finish, however, with a little speculation and prophecy.
254 TRANSACTIONS I. E. S. — PART II
I have shown elsewhere* that if we could produce white light
with no accompanying invisible radiation or other losses, by some
merger of- fire-flies, there should result an efficiency of 330 lumens
per watt, or about 40 times that of the tungsten lamp. Since
16,000 lumens is necessary to produce daylight intensity and dis-
tribution in this room, there would be required for this ideal
1 • 1-1 16,000 . ,rt, . . . .
white light , or about so watts. This, it is interesting to
6 300 °
note, is just about what the builder of the house has provided
for in the low-hung, glaring and utterly horrible central fixture.
Artificial daylight may not always be a luxury.
DISCUSSION.
Prof. George A. HoadlEy : It seems to me that Dr. Ives
has given us in the paper which is before us an excellent example
of how to carry on laboratory experimentation in our own homes.
It is true that the subject under consideration lends itself
more than some others would, to such investigation but there
are many phases of the problems of lighting that require just the
kind of semi-leisure investigation that can be carried out at
home during the long winter evenings.
Each problem that comes up in the lighting of our homes is
different from that which presents itself to our neighbor. They
all have, however, one common property, they are intensely
practical problems and if successfully solved will tend to secure
greater comfort and economy.
I fully agree with Dr. Ives in his contention that large light
sources are the most desirable. In order to secure such a source
in my dining room, I made a simple change in the fixture that
has proved to be most satisfactory. The lamp shade was of
fluted glass dark green on the outside and white on the inside,
giving a good downward diffusion. Its location was too far
from the ceiling and the light from the lamp shone directly in
the eyes of those sitting at the table. By the insertion of a ground
glass disk in the shade holder below a globe-shaped lamp, the
light comes from the area of a circle fourteen inches in diameter,
gives uniform diffusion and no glare.
* Electrical World, June 15, 1911.
SOME HOME EXPERIMENTS IN ILLUMINATION 255
Dr. C. E. Ferree : I have seen the devices described by Dr.
Ives, and I agree with him, so far as my present knowledge of
the subject goes, on all the essential points of his discussion.
The table-lamp shade seems to me to be especially good. It is
in principle, I think, the best device of its kind I have yet seen.
Among other features it possesses the advantage of shielding
the eye from the source of light without interfering with the
distribution in any other direction. A shade of this kind should
not only give excellent results for desk and table work but it
can be used very successfully for the illumination of a room.
Its degree of opacity, its distance from the light-source, and the
distance of the light-source from the reflecting walls of the
room, can all be so regulated that a fairly high degree of uni-
formity of general illumination can be obtained, — very much
more uniform than is usually obtained from fixtures of the
semi-indirect type.1 The table-lamp has, moreover, so far as
the eye is concerned, an advantage of position over the ceiling
fixture, or a wall fixture at the height at which wall fixtures are
usually placed, because it can be kept more nearly at the level
of the eye. That is, if we are to have the source of light in the
field of vision at all, it is better to have it as nearly as possible
at the level of the eye, for the image of the source when it falls
on the retina in its horizontal meridian produces less discomfort
than when it falls in the vertical meridian. When used as a desk-
light I think the shade should be made more opaque than when
used for the general illumination of a room, because when work-
ing at a desk the lines of sight are directed downwards and a
shade of the degree of transparency used by Dr. Ives would
permit more light to fall on the sensitive lower half of the retina
than should be the case if the maximum degree of comfort is
to be attained. With a shade of the dimensions used by him
probably the most favorable distribution of light over the retina
can be secured by making the shade completely opaque. The
condition to be attained is that the illumination of the retina
shall fall off more or less uniformly from center to periphery.
1 The classification of this shade as semi-indirect in type is somewhat arbitrary.
Dr. Ives may very well prefer to call it indirect since by a proper regulation of the
opacity of the shade, distance from the wall, etc.. the light in the immediately surround-
ing field may be made approximately equal in intensity to that at the source.
256 TRANSACTIONS I. E. S. — PART II
In securing this condition the factors that must be taken into
account and regulated in this case are obviously opacity and
breadth of shade, direction and distance from the eye, position
in room relative to reflecting walls, etc. The shade in principle
permits and in fact needs more or less special regulation for
each individual case.
I found the light from the inverted umbrella uncomfortable
to the eye, and from the similarity of the distribution given by
it to the distribution given by systems we have already studied,
I would conjecture that the eye would also fall off considerably
in efficiency when working under it. With regard to this light,
the following points may be noted. (1) The distribution given
by it is semi-indirect in type. In our work on distribution we
have found that the eye falls off in efficiency almost as badly
under the semi-indirect systems as under the direct. (2) It is a
ceiling light and because of the low ceilings found in most dwell-
ings it would have to be placed at a height above the level of the
eye that would be very uncomfortable. As it is installed in Dr.
Ives's home, its angle of direction from the eye is very near to
that which we have found to give the greatest discomfort. (3)
The light is rendered yellowish by transmission through the
shade. So far as our work on the effect of quality of light on
the eye has been carried, we have found greater loss of efficiency
and more discomfort under yellow light than under lights whiter
in quality. (4) Sufficient diffusion is not produced by the um-
brella as a shade to break up entirely the images of the light-
source.
Dr. Ives's photometric analysis of the distribution of light in
a room illuminated by daylight from windows is interesting and
very suggestive as to methods of attack on the problem of light-
ing. His device to reproduce this distribution for artificial light
I find, however, to be very uncomfortable to the eye. This
difference in effect on the comfort of the eye presents an inter-
esting problem for solution. Dr. Ives states that he has repro-
duced by means of his artificial window, at a lower scale of
intensity, the relative distribution of light in the room gotten
from the daylight window. That is, the ratio of intensity of
light at the source to that at various points in the surrounding
SOME HOME EXPERIMENTS IN ILLUMINATION 257
field was made the same for both cases. From the standpoint
of the light in the room, then, the only difference between the
two cases is in terms of his statement apparently in the quality
of light and in the scale of intensity used. This difference, I
understand, is believed by Dr. Ives to be the cause of the differ-
ence in the effect on the comfort of the eye. I should not myself
be inclined fully to accept this explanation until more differential
evidence is obtained. The effect of both factors in question can
be investigated under conditions in which it is more definitely
certain that no other factor is present. For example, the effect
of quality of light on the comfort of the eye can easily be tested
out by separate experiment, also the effect of changing the scale
of intensity when a definite ratio of intensity between a given
light-source and the surrounding field is maintained. I hope
that later both of these points will be investigated. On the latter
point I have at present this much evidence to offer. When
working with either a direct or semi-indirect system of lighting,
the eye both falls off less in efficiency and experience less dis-
comfort at the lower scales of intensity than at the higher. The
most favorable intensity too is found to be less than that at-
tained by Dr. Ives with his artificial window. If it can be
assumed, then, that the ratio of intensity of source to surround-
ing field in our work remained approximately constant as the
general scale of intensity was changed,2 as I have every reason
to believe it did, the above results would lead, so far as the case
is representative, to a conclusion which is just the reverse of
that suggested by Dr. Ives in partial explanation of the excess
of discomfort caused by the artificial window. I would in fact
myself be very much inclined to seek further for a factor in
the cause of the discomfort. Dr. Ives secured his distribution
by a number of reflecting plates set at different angles with
small spaces between. The reflection from these plates was,
moreover, only partially diffuse. To the eye in any given posi-
tion, the surface brilliancy of the plates was not by any means
uniform nor even uniformly graded from point to point. Quite
2 Ifi these experiments clear tungsten lamps ranging in wattage from 15 to 100 were
used. In each test the wattage from fixture to fixture was uniform, ;. e., the lamps in
all the fixtures were either 15's, 25's, 40's, 6o's, or 100's. depending on the intensity de-
sired.
258 TRANSACTIONS I. E. S. — PART II
considerable specular reflection and glare were present. Viewed
by the eye in any given position, the window had more or less
the appearance of several light-sources of different intensities.
The surface brilliancy of a window illuminated by daylight seems
on the other hand, to the unaided eye at least, to be more uni-
form, or at least more uniformly graded. The surfaces that
reflect the light into the room, corresponding to Dr. Ives's plates,
— the sky, the pavement, the walls of neighboring buildings, etc.,
are in general not in the field of vision or at least not so largely
in the field as are Dr. Ives's plates. Moreover, the reflecting
surfaces which are in general concerned in the illumination of a
room by daylight give diffuse reflection and not specular. In
short, it seems that something that affects the comfort of the eye
has escaped Dr. Ives in his photometric analysis of distribution.
Whether or not it is something that could be detected by photo-
metric analysis I am not prepared to say. It would seem to me,
however, that if an adequate check were to be had in both
cases on distribution of light in the room and of surface bril-
liancy in the source, readings would have had to be taken in a
greater number of directions than were taken by Dr. Ives. That
is, it seems to me possible to have duplicated by means of re-
flecting plates, especially plates which were not completely dif-
fusely reflecting, set at appropriate angles, the readings in the
number of directions he employed and still not have had a dis-
tribution and surface brilliancy by any means identical with that
present in his illumination by means of the daylight window.
Dr. H. E. Ives (communicated) : Upon seeing the amplified
discussion which Dr. Ferree has submitted in writing, I feel it
encumbent on me to describe more in detail some of the experi-
mental conditions and correct what I believe are misconceptions
on his part. In general Dr. Ferree believes that the third copy
of daylight described is not as complete as the paper would lead
one to believe, — that differences are present that escape the
photometer. I think that Dr. Ferree has not properly grasped
the fact that the measurements given are not of illumination.
but of surface brightness. Thus, when he speaks of the "most
favorable intensity," as determined by his own experiments, being
less than that of these experiments, he can be giving only an esti-
SOME HOME EXPERIMENTS IN ILLUMINATION 259
mate of intensity of illumination, since the intrinsic brightness
of the direct and semi-indirect units being studied by him is tens
or hundreds of times greater than the brightest visible object in
the window experiment. In fact, so extremely different in mag-
nitude and distribution are the visible bright areas in Dr. Ferree's
experiments and mine that it appears to me out of the question
to attempt any comparison based on a mere guess at the relative
illuminations. Considerable stress is laid by him on the "specular
reflection" of the plates constituting the artificial windows. My
description of these as partly specular was unfortunate, as it
gives the idea of reflected images. A better description would be
"matt reflecting surfaces of varying reflection coefficient in differ-
ent directions." The sandblasted mirrors as used give reflected
"images" of perhaps one foot diameter. The light sources were
six inches apart; consequently it is easily possible to arrange
these reflectors to present a surface entirely uniform in bright-
ness. It is physically impossible, when care is taken to secure
perfect uniformity of reflecting power in the various elements,
to distinguish by inspection from any given direction that the
reflection is not perfectly matt.
It is quite true that the different sheets were not uniformly
bright, but this was because before Dr. Ferree saw the installation
absolute uniformity was tried out and found far more trying to
the eye than a certain amount of irregularity, which was. there-
fore, allowed to creep in. This latter is decidedly less than that
present by day in the landscape seen through the window.
I. therefore, hold to my claim that the conditions produced
were a very accurate copy of the daylight conditions and appear-
ance, with the exceptions noted, namely, the proximity of the
bright surfaces, the low absolute intensity and the color of the
light. I entirely agree with Dr. Ferree tkat these factors ought
to be tested out separately to determine their relative importance.
We look to the psychologist to do this.
Mr. Robert B. Ely : Decorators have, in numerous instances,
used cretonne screens on candles in dining rooms and bed rooms,
etc., so constructed as to reflect the light on the wall, the shades
being elliptical in shape and so placed as to conceal the light
260 TRANSACTIONS I. E. S. — PART II
source from view. In a great many of these instances I have
found that there has been complaint of poor illumination in the
center of the room with the candles equipped in this manner.
In bed rooms this has been overcome by placing two small port-
able lamps equipped with shades on bureaus or chiffoniers, so
that the light was reflected from bureau scarf to illuminate the
face of the person standing in front of the bureau or chiffonier.
In dining rooms it has been necessary to resort to the use of a
candelabra, equipped with imitation candles and suitable shades.
Mr. R. L. Lloyd: It seems to me the natural direction for
light is from overhead. We are living under artificial conditions
in houses with windows in the side walls, and have become ac-
customed to seeing light enter that way, but the natural tendency
will generally be found to be to turn one's back to the light, so
as to keep it out of the eyes. When the Ancients first began to
utilize artificial light they turned naturally toward locating it
above, and I understand that some of their temples were made
without roofs, so that light could enter in the natural direction.
Although these experiments of Dr. Ives are valuable as re-
searches in science, I think he is working in the wrong direction,
in trying so assiduously to imitate lighting as entering from
windows, when it would be much more easily accomplished to
arrange the light to come from above. We have seen by the
demonstrations of Mr. Luckiesh and others that when the light
is directed on natural objects from above, we see them in their
natural appearance, and that when the light is directed from
other than above, their appearance is distorted. The photo-
grapher too has learned this, and always makes use of a skylight
in his studio for taking pictures.
As stated before, these experiments of Dr. Ives are very
interesting and are valuable contributions to our science, but I
have yet to be convinced that a simulation of artificial conditions
is a proper one for the best results.
Prop. A. J. Rowland: I have had the privilege of seeing the
installation of lights and lighting referred to in Dr. Ives's paper.
I must confess to a certain first sense of bewilderment when I
saw Dr. Ives's installation, especially that for producing day-
SOME HOME EXPERIMENTS IN ILLUMINATION 26l
light values. I am so used to seeing lights in certain places,
mounted in a conventional way, that I have pretty strong feel-
ings about something special or unusual. The simplicity of the
side wall lighting scheme, and the good results secured from it,
impressed me strongly. When it can be used intelligently, it
seems to me to be one with points of great merit. The trouble
with most plans which secure good lighting by the use of things
outside the lighting fixture itself as a source, comes from the
extraordinary way in which applications are made by light users.
Dr. Ives's walls are covered in such a way that they lend them-
selves splendidly to use as secondary light sources. At my home
a deep green felt paper would make the plan worthless. I
wonder whether in ordinary service most people would not
swing the shade 180 degrees from its proper position and utterly
spoil the lighting scheme. I like the large source plan; I like
the idea of placing the main sources of light on levels similar
to those from which daylight is derived; and the simplicity of
the whole side lighting plan has much to commend it.
In connection with the paper, the term "brightness" has been
used a number of times. I think it is one which needs to be
explained carefully. If one compares the brightness of a white
wall a gray wall, or a dark colored floor, it is a thing independent
of color. Just what does it imply? How is it to be measured?
Mr. Charles O. Bond: The semi-indirect side-wall lighting
scheme proposed and favored by Dr. Ives deserves particular
attention. He shows that strong contrasts in the field of vision
are avoided ; that the method lends itself to artistic treatment and
adornment of a room into which even color variations may be
introduced ; and, best of all, the method has "worn well."
His data have been obtained in a home, and in such conditions
fixed positions are not compulsory. If the lighting is found try-
ing while one is occupying one seat, it is easy to change to
another. That is one reason why such indefensible lighting pre-
vails in many houses.
The side-wall semi-indirect method ought to be easily adaptable
with greatly increased resultant comfort to small audience cham-
bers, such as church parlors, where persons may not easily change
the lighting conditions through a change of seat.
262 TRANSACTIONS I. E. S. — PART II
Dr. H. E. Ives (In reply) : I have nothing to add to the
discussion. My hope is that this paper may direct thought to
certain other possibilities in lighting than are now— due to effi-
ciency considerations — most common.
PIERCE: GASLIGHTINC IN AN EXHIBITION HALL 263
GASLIGHTING IN AN EXHIBITION HALL.*
HV ROBERT F. PIERCE.
Synopsis: The following article describes a temporary semi-indirect
gas lighting installation which was provided for an exhibition hall. Ten
lighting units, each consisting of fifteen upright burners within an orna-
mental glass bowl mounted on a pedestal, furnished the illumination. The
illustrations show the plan of the unit, a diagram and a night view of
the interior.
The illumination of Taft Hall in the Auditorium Armory,
Atlanta, Ga., is the result of a number of compromises with
more or less unfavorable elements, rather than the unrestrained
working out of a consistent and coherent plan, and must be
judged in the light of the existing circumstances. The lighting
was primarily designed for use in connection with the conven-
tion of the National Commercial Gas Association. The same
room was to be used for the beefsteak dinner with which the
convention was terminated, and, in addition to this, it was desired
to furnish a lighting system which should be available for the
automobile show which preceded the gas convention.
The problem then resolved itself into meeting, as well as pos-
sible, the requirements imposed by each of these widely varying
purposes with one installation. Naturally, it was desired that the
installation embody unique features— at least, features unique as
far as gas lighting is concerned — and it was suggested that indi-
rect lighting be employed in connection with pedestals ; something
along the line of the installation in the Louis XVI dining room
in the Congress Hotel, Chicago.
It so happened that the room was peculiarly adapted to the
support of the lighting units on columns or pedestals. In fact,
this was the only feasible arrangement. The ceiling is plain and
unbroken, except by four beams which divide off a large square
in the center of the ceiling comprising nearly one-half the ceiling
area. This square is further sub-divided by a single longitudinal
beam of the same dimensions as the others. Obviously the ceil-
ing presented no location for the suspension of fixtures.
The side walls were devoid of pilasters or similar features to
* A paper read before a meeting of the New York section of the Illuminating: Engi-
neering Society, January 9, 1913.
5
264 TRANSACTIONS I. E. S. — PART II
which wall brackets might have been attached, and the only
remaining expedient was in the use of columns for the support
of the lighting units.
For the lighting of the hall for convention purposes, the best
arrangement of the columns appeared to be as shown in Fig. 5,
marking off an area bounded by the supporting columns and
fronting upon the rostrum from which the convention was con-
ducted. This arrangement also served very well for the lighting
of the automobile show, the booths being arranged in such a
manner that the lighting columns marked the rear corners of
the booths when viewed from the middle aisle, and the front
corners when viewed from the side aisles.
In the design of the lighting columns, it was considered desir-
able to depart from the Ionic order of the supporting columns for
two reasons. In the first place, the use of similar columns would
have given them the appearance of being original structural
members, decapitated for the purpose of bearing the lighting
units, whereas, it was felt, a different treatment would separate
the lighting columns into a distinct system, having its own reason
for existence. In the second place, it was desired to utilize a
design which would not be out of place in other surroundings,
as it was likely that the installation would be sold for other pur-
poses after the convention. The composite Corinthian order was
selected as having the widest field of possible future applications
and one which would harmonize fairly well with any classic
interior.
In the design of the light distribution the direct, semi-indirect
and indirect systems received consideration. The latter, origin-
ally suggested, had a number of serious drawbacks. Any system
of illumination which turns the room optically upside down by
reason of reversing the natural order of intensities increasing
from the ceiling downward, is certain to be a source of more or
less definite annoyance, especially where the attention of the
occupants of the room is not apt to be concentrated upon work
which distracts their attention from the abnormal distribution of
illumination in the room. This objection might be of little
moment in a room devoted to clerical purposes, but in the hall
under consideration, was held to be of much importance. The
*/>
Fig. i. — View of interior of hall.
Fig. 2.— Design of bowl
and pedestal.
Fig. 3. — Plan of light unit, showing
arrangement of lamps.
Fig. 4. — Distribution curve from light unit. In the above distribution curve
the radii vectores of the right-hand curve represents the fluxes in zones ex-
tending 5 degrees on either side of the designated angle, while those of the
right-hand curve represent the total flux from zero to the designated angles.
Fig. 5. — Plan of interior of the hall.
PIERCE: GASUGHTING IN AN EXHIBITION HALE 265
absence of an apparent source of light was deemed a considerable
drawback to the indirect system from a similar consideration.
Furthermore, only the presence of a luminous light source could,
from an esthetic standpoint, provide any excuse for the existence
of the lighting columns.
For the lighting of the automobile show it was felt that the
presence of luminous light sources was especially desirable, as
the appearance of an abundance of light seems to be an indis-
pensable feature of exhibitions of this kind.
The foregoing considerations seemed to narrow the choice to
the selection of either direct or semi-indirect lighting.
For the automobile show, an illumination of about 8 lumens
per square foot was deemed desirable on account of the light-
absorbing qualities of the black car-bodies. The area of the
hall (about 5,900 square feet) demanded, therefore, a horizontal
component of about 47,000 lumens, which demanded, at a utiliza-
tion efficiency of, say 40 per cent, (the walls are dark green in
color) about 120,000 lumens generated. It seemed desirable to
limit the number of lighting units to ten in order that they might
not interfere with the best use of the floor space. This demanded
the production of 12,000 lumens per unit. One candle-power per
square inch was decided upon as an upper limit for the specific
intensity of the bowls or globes. This would have required a
globe of a size altogether out of proportion with the columns,
and it was finally decided to use semi-indirect lighting.
An indirect component of 50 per cent, was decided upon as
meeting the various requirements to the most satisfactory degree.
This ensured the reduction of shadow contrasts to a point well
within the requirements of the automobile show, where ample
diffusion of light for the inspection of machine interiors was
required and at the same time avoided the "flatness" which would
have accompanied a much lower indirect component. It was
believed that the 50 per cent, direct component would be sufficient
to bring the apparent plane of highest . illumination well down
toward the floor, where it belonged.
A light distribution (Fig. 4) was finally decided upon having
an extensive character above the horizontal, with a maximum
at 135 degrees. With the units as located, this provided sub-
266 TRANSACTIONS I. E. S. — PART II
stantially uniform ceiling illumination, preserving the flatness of
the ceiling. The ceiling being a very light cream in color, it was
assumed that about 50 per cent, of the light from no degrees
upward would be effective on the working plane, giving about
30,000 effective lumens from the ceiling. Assuming 70 per cent,
as the effective angle below the horizontal, 30,000 lumens would
be contributed to the working plane direct from the lamps, giving
an average effective illumination of about ten lumens per square
foot — about 50 per cent, of which is indirect.
The bowls were primarily designed to furnish this curve by
reflection from the glass itself, and to harmonize the form of
the bowl with the column, it was treated as a conventionalized
floral form, borne by the conventionalized tree trunk and foliage
which comprised the shaft and capitol of the columns. For this
reason, it was necessary to depart from the strictly classic orders
in designing the bowl. Equalite glass was selected for the bowl
on account of its shell-like texture and selective absorption, giving
a warm lively glow in the more transparent of the irregular stria-
tions of its structure. This relieved the cold, severe white of the
denser portions without conflicting with the material of the col-
umns, which was a concrete composed of cement and marble dust.
The necessity of avoiding colors in the bowl itself imposed the
use of conventionalized floral forms for decoration which should
be devoid of any suggestion of color, and the lotus was selected
for this purpose as harmonizing both in form and color with the
bowl. This might be criticised as an injudicious admixture of
Greek and Egyptian orders, but it was felt that such criticism
is captious, there being no necessity for adhering to purely
geographical distinctions in this instance. The design of bowl
and column finally selected is shown in Fig. 2.
Fig. 1 was made from a photograph of the completed installa-
tion and conveys a fair idea of the distribution of illumination.
It appears from this picture that the indirect illumination is
rather overdone. The dark floor, however, contributed a great
deal toward this impression.
It was exceedingly unfortunate that the ceiling fixtures and
draperies could not have been removed. They were extremely
offensive and marred what would otherwise have been a quite
PIERCE: GASLIGHTING IN AN EXHIBITION HALL 267
agreeable interior. It was also unfortunate that the interior could
not be redecorated in such a way as partly to correct for the inver-
sion of illumination intensities by a compensating color gradation.
Had this been possible, however, it is doubtful if it would have
availed much with the high intensities made necessary for the
lighting of the automobile show.
Fig. 3 shows the arrangement of lamps in the bowls. Fifteen
upright burners are mounted in two concentric rings and supplied
with gas through magnet valves. Ignition is accomplished by a
flash pilot, throwing a jet under an inverted annual trough above
the burner chimneys. This pilot is simply an automatic open-
flame lighter, or Boston cock, which is lighted by a make-and-
break spark — an arrangement familiar to everyone who has used
open-flame gas burners.
The entire burner arrangement is composed of standard parts
and appliances which have been used for many years with entirely
satisfactory results, so that this installation is in no sense a special
one without commercial utility, but exemplifies possibilities in the
way of securing unique and pleasing effects which may be utilized
by any designer of gas lighting installations who cares to avail
himself of them.
268 TRANSACTIONS I. %. S. — PART II
METAL REFLECTORS FOR INDUSTRIAL LIGHTING.*
THOMAS W. ROIvPH.
Synopsis: After brief introductory comments on the progress in
industrial lighting, the author of this paper presents a classification of
metal reflectors and a discussion of the nature of reflection. For the
most part, the paper is devoted to the merits and demerits of porcelain-
enamel and aluminum-finished steel reflectors and a consideration of the
distribution characteristics which influence the selection of the different
types of these reflectors for industrial lighting installations. The cost
of reflectors, the author states, is of minor importance in that it is
readily compensated for by effective illumination when the lighting system
is properly designed.
Up to the present time, the progress of industrial lighting has
followed the progress of lighting in the general field. In effi-
ciency, in variety and excellence of distribution, in diffusion and
in eye-protection, the improvement of reflectors for use in the
general lighting field has preceded the improvement of industrial
lighting reflectors. We have now reached a point at which this
condition is likely to change. Industrial lighting is assuming a
greater importance and it is not at all improbable that some of
the most noteworthy advances in illuminating engineering in the
next few years will originate in this field. The reason is not
hard to find. This is a commercial age and whatever can be
reduced to dollars and cents will receive the maximum amount
of attention from the business men of the country; wherever im-
provements can be shown to affect profits most vitally, there im-
provements will be most rapid. In the lighting of factories, good
lighting — meaning adequate intensity of illumination, proper
protection of the eyes and a high illumination efficiency — can be'
shown to affect profits more directly and to a greater degree than
in the general field of commercial lighting.
Those interested in the advancement of industrial lighting
have before them the task of disseminating among a large body
of variously employed individuals the underlying principles of
good illumination. Factory engineers and managers, who as a
class have already shown their appreciation of good lighting,
* A paper read at a meeting of the Philadelphia section of the Illuminating fingineer-
ing Soeiety, May 16, 1913.
R0LPH : METAL REFLECTORS FOR INDUSTRIAL LIGHTING 269
salesmen who sell all kinds of reflectors, bad as well as good, gas
and electric companies' solicitors, who are expected to sell results
in illumination as well as gas or electricity ; all these should have
a clear understanding of what constitutes good illumination and
of how reflectors control light in producing it. The proper redi-
rection of the light of the lamp by means of reflectors is essen-
tial to good lighting. A knowledge of the merits and demerits of
the reflectors available for factory lighting is, therefore, of par-
ticular importance.
Metal reflectors for industrial lighting may be classified as
follows :
f Material
Character of
Reflection
Steel
Brass
Aluminum
( Specular
Spread
Metal
Reflectors
L
Diffuse
•( Polished metals
f Rough metal
surfaces
Applied
aluminum
Porcelain
enamel
Paint enamel
Shape
Distribution of
Light
f Deep bowl
j Shallow bowl
] Shallow
^ Angle
Extensive
Intensive
■{ Focusing
I Distributing
L Asymmetric
The metal most widely used, and rightly so, is steel. It is
durable, reasonably low in price and readily takes an applied
finish of almost any character. Brass and aluminum are both
more expensive than steel and somewhat less durable. Brass
finds a limited field of usefulness in reflectors of such shapes that
cannot be drawn or spun of steel. Aluminium has a slight ad-
vantage because it is light in weight and its surface, when not
polished, is excellent from the standpoints of efficiency of re-
flection and character of illumination produced.
The character of reflection obtained from a metal reflector
is very important in determining the value of a reflector. By
270 TRANSACTIONS I. E. S. — PART II
character of reflection is meant the character of the action of the
surface upon each minute pencil of light-rays, i. e., whether it re-
flects the pencil regularly changing its direction only, or whether
it breaks up the pencil, reflecting light in many different direc-
tions. There is a very general lack of understanding of these
actions. The law that the angle of reflection is equal to the angle
of incidence, i. e., that a light-ray is always reflected at the same
angle with the surface as the angle at which it strikes (Fig. i-a),
is given credit for a much wider field of usefulness than it really
has. This law holds for every reflection of an individual light-
ray by an infinitesimal portion of a surface ; but many surfaces
are rough and many other surfaces allow light to pass into them
and be reflected from particles beneath, so that light-rays which
are parallel when striking a surface, are often broken up and
reflected in many directions.
Reflecting surfaces in common use may be grouped into three
distinct classes, according to the manner in which they reflect
light. These three varieties of reflection are specular reflection,
spread reflection and diffuse reflection. Fig. 1 shows the char-
acter of each of'these. In each case, the dotted line surrounding
the reflected rays, is what might be termed the photometric curve
of the light reflected from any point on the surface. Specular
reflection rigidly follows the law that the angle of reflection is
equal to the angle of incidence. The reflecting surface is smooth
and the reflected ray always makes the same angle with the sur-
face as the incident ray. In spread reflection the maximum
candle-power of reflected light is in the same direction as in
specular reflection. The light is broken up, however, and slightly
spread from the direct path. In diffuse reflection the angle of'
incidence has no effect upon the reflected light. No matter at
what angle the light strikes the surface, the maximum reflected
ray is normal to the surface and the light is reflected in all
directions in accordance with the well-known cosine law. The
photometric curve of each point on the surface is a tangent
circle. Fig. 2 illustrates the manner in which each of these three
kinds of reflection is produced by the infinitesimal portions of the
surfaces. Regular reflection is produced by any smooth opaque
surface. Mirrors and polished metals produce regular reflection.
\r
tp,*
Fig. i.— Character of reflection obtained from various surfaces,
(a)— Specular reflection; (b)— spread reflection; (c)— diffuse reflection.
o d ° ° -
Fig. 2. — Reflecting action of various surfaces.
a Regular reflection; (b) — irregular reflection; (c) — sub-surface reflection.
Fig. ?. — Combination of diffuse and specular reflection.
s
S SSS3
s
s s
1 ( 1 1 1 lnr~~j^M
JpX~^\ \ \-\-X\
' Ul
Xt _-'--•! o-
c
LJ/ J 1 1
>CJ~8i0""
)*OpriZ0-
v^T"°-i60-
ciomjjAlone X/
-Lompw.fhR.rflecJpr^
Fig. 4. — Extensive aluminum finished
steel reflector. (Photometric curve
shown in Fig. 5.)
Fig. 5. — Photometric curve of extensive
aluminum finished steel reflector, with
100-watt clear tungsten lamp operating
at 1. 13 watts per candle. (Reflector is
shown in Fig. 4.)
Fig. 6.— Shallow type distributing steel
reflector, porcelain enameled. (Photo-
metric curve shown in Fig. 7.)
Fig. 8. — I,arge size angle reflector, porce-
lain enameled steel. (Photometric
curve shown in Fig. 9.)
j — l y^v\>^ \
''1 TiujIl
/ '.
1 1 M
1 v^*-"***!^ '■-■*•'•' .
xf\y~cr/s" 1^0 — y
Vv^Cj^SkT"
Vv
^\ ?• — 300
ti«y^/
Fig. 7. — Photometric curve of shallow
type of porcelain enameled steel re-
flector with 150-watt clear tungsten
lamp operated at 1.13 watts percandle.
(Reflector shown in Fig. 6.)
' v V'V"-
d$
AXV
VLar
np Alor.e\^\.--
injMthRefledor
V^f~7
x>3v 40° \
///f*^l
^ 75—800 -
^^T 1000 -
<x//
150°
Fig. 9.— Photometric curve of large size
porcelain enameled angle steel reflec-
tor with 500-watt clear tungsten lamp
operating at 1.00 watt per candle. (Re-
flector shown in Fig. 8.)
ROLPH : METAL REFLECTORS FOR INDUSTRIAL LIGHTING 2J I
Irregular reflection of the individual light-rays (Fig. 2-b) causes
a spreading of the light, resulting in spread reflection. This is
produced by any matte surface. The surface of this character
most widely used for reflectors is the applied aluminum finish.
Diffuse reflection is almost invariably produced by reflection of
light-rays from particles beneath the surface (Fig. 2-c). The
most widely used surface of this character is porcelain enamel.
Porcelain enamel is, in reality, a glass having the characteristics
of opal. The minute particles held in suspension in the glass
are the reflecting media. In addition to the light reflected from
these particles beneath the surface, there is a small amount of
reflection from the surface itself. This reflection is regular since
the surface is smooth. The result is a combination of diffuse
and specular reflection as shown in Fig. 3. The light reflected
specularly is usually negligible in quantity, but cannot be entirely
overlooked, especially in reflector design.
Surfaces giving specular reflection afford a high degree of
light control and polished metal reflectors could be designed to
give almost any desired distribution of light. Such surfaces are
very little used, however, due to the streaked character of the
illumination obtained from them. The polished aluminum re-
flector, used considerably several years ago. is now well-recog-
nized as a highly undesirable reflector because of the striations
or streaks in the resulting illumination. Surfaces giving spread
reflection are only slightly less susceptible to accurate design than
surfaces giving specular reflection since with spread reflection,
the major portion of the reflected light deviates only slightly
from the law that the angle of reflection is equal to the angle of
incidence. Aluminum finished reflectors, for example, can be
accurately designed to give all the most widely useful kinds of
light-distribution. Diffusely reflecting media, however, such as
porcelain enamel and paint enamel, have a much narrower field
of usefulness. Extreme distributions pf light, such as the focus-
ing type, cannot be obtained with reflectors of this character.
The law of diffuse reflection is a law which can be accurately
applied to reflector design, but the possibilities of obtaining va-
ried distributions of light with diffuse reflectors are exceedingly
272 TRANSACTIONS I. E. S. — PART II
limited. The users of reflectors and even the manufacturers in
many cases, have quite generally failed to recognize this.
Referring again to the classification of metal reflectors, it
will be seen that in shape there are four general classes — deep
bowl, shallow bowl, shallow and angle. Deep bowl, shallow
and angle types of reflectors are well known. The shallow bowl
has been less generally used. Figs 4, 6, 8 and 10 show typical
reflectors of these four classes. The deep bowl and the angle
types (when the latter are rightly used) are much preferable
to the others from the standpoint of protecting the eyes from
exposure to brilliant lamp-filaments. The shallow bowl and
shallow reflectors, however, are the only shapes giving a dis-
tributing type of curve and this curve has certain fields of use-
fulness. As a subclassification of shallow reflectors, the flat
type might be mentioned, although these are rapidly becoming
obsolete. Flat reflectors usually allow the lamp filament to pro-
trude too far below the lower edge of the reflector. The shallow
reflectors now most commonly in use are sufficiently deep to
cover the lamp at least to the bottom of the filament.
Distribution of light is perhaps the most useful method by
which reflectors can be classified. Certainly, in selecting the re-
flectors for use in any installation nothing is more important than
to obtain the best distribution of light for the purpose. Exten-
sive, intensive and focusing distributions are well-known and
widely used. The prototype curves of these distributions (i e.,
the original curves which reflectors should be designed to give)
are calculated on the basis of obtaining uniform illumination
when the light-units are arranged in squares with the distance
apart bearing a definite ratio to the height above the plane of
illumination. These ratios of distance apart to height are 2 for
the extensive distribution 1% for the intensive distribution and
% for the focusing distribution. Fig. 4 shows an illustration of
the well-known extensive reflector and Fig. 5 shows its photo-
metric curve; Figs. 12 and 13 show the illustration and curve
of an intensive reflector and Figs. 14 and 15, a focusing reflector.
Reflectors most nearly approaching the prototype curves have
been selected. Reflectors giving these distributions are available
for inverted gas and for nearly all the various sizes of tungsten
ROLPH I METAL REFLECTORS EOR INDUSTRIAL LIGHTING 273
lamps. Extensive and intensive reflectors are available in both
aluminum and porcelain enamel finish, although reflectors hav-
ing porcelain enamel finish cannot be designed to give as good
an extensive distribution as reflectors having an aluminum finish.
Focusing reflectors are not available in porcelain enamel since,
as stated above, a focusing distribution cannot be obtained with
a diffusely reflecting surface.
The distributing type of photometric curve is characterized
by high candle-power values at the angles of 500 to 750 from
the nadir or vertically downward direction. To obtain high can-
dle-power values at these angles it is necessary to expose the
lamp filament considerably more than in the case of reflectors
giving extensive, intensive and focusing distributions. Conse-
quently, while deep bowl reflectors may be used to obtain ex-
tensive, intensive and focusing curves, shallow or shallow bowl
reflectors must be used when a distributing curve is desired.
Figs. 7 and 11 show good distributing curves obtained from a
shallow porcelain enameled reflector and a shallow bowl alumi-
nized reflector respectively. This radical difference in shape of
reflector to obtain the same type of distribution is due to the
radical difference in the character of the reflection obtained from
porcelain enamel and aluminum, as explained above. Shallow
reflectors if made with an aluminum finish would not give a dis-
tributing curve unless the lamp filament were allowed to project
considerably below the edge of the reflector, thus wasting en-
tirely too much light at and above the horizontal. Fig. 16 illus-
trates this, by showing the photometric curve of a shallow type
of reflector in porcelain enamel and the curve of the same re-
flector in aluminum finish. It will be seen that the change in
finish has changed the character of distribution entirely. The
aluminum finished shallow reflector has little practical value. It
is not distributing, while as a focusing type it has too high candle-
power values at 500 to 750 from the vertical, and does not protect
the eyes from the lamp filament to as great a degree as the regular
types of focusing reflectors. When a distributing curve is de-
sired in aluminum finish the shallow bowl type of reflector should
be used. This shape was developed solely to obtain a distrib-
uting: reflector in aluminum finish.
274 TRANSACTIONS I. E. S. — PART II
Distributing reflectors have been more generally used than
they should have been. As stated above, their principal charac-
teristic is high candle-power values at angles of 500 to 75 ° from
the nadir. To obtain these high candle-power values a certain
degree of eye-protection is sacrificed. Consequently they should
be used only where the light at these high angles is of more im-
portance than the better eye-protection which would be obtained
with deep bowl reflectors. There are certain cases where light at
high angles is of importance. For example, when illumination
is required on many different vertical or oblique surfaces, high
above the floor, the distributing reflector will often prove the best
reflector to be used. It frequently happens, however, that such
cases are better taken care of by the proper type of angle reflec-
tor. As another example, large stock-rooms and warehouses
should usually be lighted with distributing reflectors. Here lit-
tle actual work is performed and exposure of the lamp filament
is permissible, while to obtain a reasonably low cost of installa-
tion, the light-units must be placed far apart. When the distance
apart is greater than two and one-half times the mounting height,
extensive reflectors will leave dark spots half-way between light-
units. Distributing reflectors will eliminate these dark spots.
It must not be supposed, however, that distributing reflectors will
give uniform illumination at wide spacings. Their advantage
lies simply in the fact that at these wide spacings, they do not
allow the intensity of illumination half-way between units to
drop as low as other reflectors do. To obtain uniform illumina-
tion, extensive reflectors can be used farther apart than any other
type of metal reflectors on the market. The spacing constant for
uniform illumination with extensive reflectors is k = 2. i. e., dis-
tance apart should be two times the height above the plane of
illumination. The distributing curve does not give uniform illu-
mination unless the spacing is k = 1.6 or less, thus requiring
units considerably closer together than when the extensive distri-
bution is used. This does not mean that distributing reflectors
should be installed at the spacing k = 1.6 or distance apart 1.6
times the mounting height. When uniform illumination is de-
sired extensive, intensive or focusing reflectors should be used.
7
Fig. 10.— Shallow bowl type of distribut-
ing reflector, aluminum finish. (Photo-
metric curve shewn in Fig. ii.j
I
^
w r^
5r-
v \ j<\\ vvSO^ *° ~^//
////
\ v >c^^. 7\ /"t^ — ^
V/>)
vxxNc^TT""*2 _^-'
w so»
Fig. ii.— Photometric curve of shallow
bowl type distributing reflector, alumi-
num finish, with 60- watt clear tungsten
lamp operatingat i.i6watls per candle.
I Reflector shown in Fig. 10.;
> c
Fig. 12. Intensive aluminum finished
steel reflector. 'Curve shown in Fig.
130
Fig. 13.— Photometric curve of intensive
aluminum finished steel reflector with
100-watt clear tungsten lamp operating
at 1. 15 watts per candle. (Reflector
shown in Fig. 12.;
I reusing aluminum finished
steel reflector. (Photometric curve
shown in Fig. 15.;
Fig. 15.— Photometric curve of focusing
aluminum finished steel reflector with
100-watt clear tungsten lamp operat-
ing at 1. 13 watts per candle.
Fig. 16. — Photometric curves of shallow dome type reflector,
porcelain enamel and aluminum finished with 250-watt
clear tungsten lamp operating at 1.00 watt per candle.
Fig. 17. — Small angle type of reflector,
aluminum finished. (Photometric curve
shown in Fig. iS.)
li^y^' ^fei
\^\^C\\\^\7^r^\^^^I~/rCr~l~~L
W \ \-Ax^4yT\3<v/Sc A / /
\\AXX^S5^/XA0
\xxSsv\j ^ v° — X^\//>
\\^\y\Cr/~ — ^K~~~C^^j<//
/>
vva^T^^-IIZ-6!0 "ZSA^^-a^/
"^ LampwithReftecTSF — 70 ^jSx^^V
Fig. iS. — Photometric curve of small
aluminum finished angle steel reflector
with 25-watt clear tungsten lamp oper-
ating at 1.1S watts per caudle.
-A" r ./^~
\
P055IBLE
™z<
■ -'V
_LJ_N E_0 F X1 *i5>ii
Fig. 19. — Diagram showing angle above which light emitted from
light-units is likely to cause eye-strain.
ROLPH : METAL REFLECTORS FOR INDUSTRIAL LIGHTING 2/5
Distributing reflectors are for wide spacings where uniform illu-
mination is unnecessary.
The fact that distributing reflectors are less undesirable than
extensive with very wide spacing of light-units causes their use
in many cases where closer spacing and extensive or intensive
reflectors would be preferable. Often a false idea of economy
dictates too small a number of outlets and distributing reflectors
are the only recourse. The result is non-uniformity of illumina-
tion and lack of diffusion. Shadows are too dense and light
comes from the wrong direction for many of the workers. The
competition among reflector salesmen is frequently responsible
for this. It is naturally easiest to sell the reflector which can be
used at the widest. spacings since the installation cost is then the
lowest. Again, the old fallacy that light is more important than
illumination is responsible for the use of many shallow reflectors.
In spite of the great spread of illumination knowledge in the last
few years, there is still a general impression that lamp filaments
must be in plain sight to obtain the best results. This tends to
increase the sale of shallow reflectors at the expense of the more
desirable deep bowls. A still more widespread knowledge of
the principles of good illumination will gradually remedy this
condition. Industrial lighting is still passing through the stage
which the general commercial field passed some years ago. Shal-
low types of prismatic reflectors are practically obsolete; shallow
opal reflectors are rapidly becoming so, but shallow metal re-
flectors are still used in large quantities. It is true that with an
opaque reflector and in a field of such diversified requirements
as the industrial field, there is some need for a shallow reflector,
but it must be admitted that the use of this type is much more
general than it should be.
The last class of reflectors given in the table under the classi-
fication "Distribution of Light" is the asymmetric class. Asym-
metric distributions are obtained from angle reflectors. These
reflectors are usually made up of symmetrical reflector forms
with the holders set at an angle. They are available in aluminum
or enamel finish for all sizes of tungsten lamps from 25-watt or
smaller to 500-watt. The small sizes are used principally for
local lighting such as the lighting of benches or particular por-
276 TRANSACTIONS I. E. S. — PART II
tions of the work on machines. They are made to give various
distributions. They can be obtained in the sizes up to 8 inches
(20.32 cm.) in diameter of such shape that the maximum candle-
power is given in any direction desired from directly downward
to directly horizontal. Most conditions, however, can be met by
a single line of reflectors, i. e., one angle type only for each size
lamp. For the small sizes, used principally for local lighting, the
greatest candle-power values should be between 150 and 45 ° from
the vertical. For the larger sizes, the greatest candle-power
values should be between 35 ° and 8o° from the vertical. It
should be noted that the angle at which the holder is set does
not indicate the direction in which the maximum candle-power
is obtained. It is characteristic of these reflectors that they give
the greatest candle-power and greatest flux of light at angles
somewhat higher than the angle at which the holder is set. Fig.
17 illustrates a typical small angle reflector and Fig. 18 its pho-
tometric curve. A typical large angle reflector is shown in Fig.
8 and its photometric curve in Fig. 9. Such reflectors are not
used for local lighting to as great an extent as the small re-
flectors. They are of value for general illumination, when there
are vertical or oblique surfaces to be lighted at a considerable
height above the floor. They are also widely used for the gen-
eral illumination of spaces in which it is difficult to place sym-
metrical reflectors effectively. Many shops, for example, have
traveling cranes which either interfere with the placing of re-
flectors above them or require such reflectors to be placed so high
as to lose much of their effectiveness. In such cases the large
angle reflectors, placed on the upright girders beneath the crane,
may be used to advantage.
The above treatment of reflectors according to their various
characteristics does not include several important features which
must be considered in selecting a reflector for any given service.
In the classification which has been made, eye-protection has
been considered only slightly, in connection with the shape of the
reflector. Efficiency, depreciation and cost have not been consid-
ered. The classification gives a broad general idea of the char-
acteristics of metal reflectors ; but for an intelligent selection of
reflectors for any given installation the following points should
R0LPH : METAL REFLECTORS FOR INDUSTRIAL LIGHTING 277
be considered: (1) eye-protection; (2) distribution of light;
(3) efficiency; (4) depreciation; (5) cost.
The above order in which these points are given will, in many
cases, be the order of their relative importance, but this will vary
somewhat, depending upon the character of the building. These
five considerations will be treated briefly from the standpoint of
their influence on choice of reflector.
The proper protection of the eyes, as far as is known at
present, involves the avoidance of three undesirable features,
namely, high candle-power, points of high intrinsic brilliancy and
extreme contrasts of brilliancy within the ordinary range of vis-
ion. When work is actually being performed the worker is usu-
ally looking in a downward direction. Wherever this is the case,
only the most flagrant violation of the principles of good lighting
(such as placing a bare lamp in front of the worker and close
to the work) will cause serious eye-strain. However, when the
worker glances up from his work, if he encounters conditions
causing eye-strain, the eyes are temporarily rendered less efficient
and it may be some minutes after he looks down again, before his
vision is normal. It is desirable, therefore, in factory lighting to
minimize the possibility of eye-strain for all ordinary conditions
of vision. The first requirement for this is to place light-units as
high above the range of ordinary vision as is consistent with good
distribution and diffusion of illumination. In large rooms, dis-
tant light-units, even if placed high, will nearly always be in the
range of vision and it is important that the units themselves be
so designed as to avoid insofar as possible, the conditions which
lead to eye-strain. Research has shown that high candle-power
ceases to affect the efficiency of the eye, when it is removed to
25 ° from the direct line of vision. It is reasonable to suppose
that high intrinsic brilliancy and strong brilliancy contrasts also
have little effect when the light-source is so far removed from the
line of vision. Assuming that the direct line of vision will not
often be above the horizontal, the requirements of light-units to
obtain best eye-protection are that the candle-power, intrinsic
brilliancy and brilliancy contrast be low at angles above 25 °
below the horizontal or 65 ° from the vertical. This requirement
is lenient since in many shops the eye is quite frequently directed
2/8 TRANSACTIONS I. E. S. — PART II
above the horizontal. However, the injurious effect of light
within this angle of 25 ° varies with the angle. The greater the
angular distance of the light from the line of vision the less will
be the injurious effect. The effect, therefore, becomes quite
small when the angle is only slightly less 25 ° from the line of
vision.
The suppression of candle-power above 65 ° from the vertical
is satisfactorily accomplished in the deep bowl reflectors of the
extensive, intensive and focusing types. The suppression of
intrinsic brilliancy at these angles requires the screening of the
filament down to 65 ° from the vertical. Most deep bowl reflec-
tors do not meet this requirement exactly ; the screening angle
varies from 650 to 750. Nevertheless, deep bowl reflectors are
reasonably satisfactory in this respect and are, of course, much
preferable to the shallow bowl reflectors. The avoidance of
strong brilliancy contrasts above 650 from the vertical, is not
always easily accomplished. When deep bowl reflectors are used,
there is often a considerable degree of contrast between the
brilliant interior of the reflector and the dark upper portion of
the room. Where these reflectors are used in rooms having light
colored ceilings and the work is principally on light colored
goods, this strong contrast is not so much in evidence. In many
cases, however, there is practically no ceiling and frequently
what ceiling there is, is dark in color. In order to relieve this
contrast, it is important that the interior of the reflector have as
low a brilliancy as possible in the directions above 65° from the
vertical. If the character of the reflection obtained from an
aluminum surface and from a porcelain surface is carefully con-
sidered, it will be seen that this has considerable effect upon the
appearance of the interior of the reflector. Aluminum reflectors
act by spread reflection. In other words the maximum candle-
power of the reflected light is at the same angle with the surface
as the incident light. In the design of these reflectors, their
contour is made of such a shape that this feature is utilized in
directing the light into the directions where maximum candle-
power is desired. In the extensive, intensive and focusing dis-
tributions, maximum candle-power is desired below 55 ° from the
vertical. Consequently, the interior of the aluminum finished
KOLPH : METAL REFLECTORS FOR INDUSTRIAL LIGHTING 2^9
reflector appears brightest when viewed from below 55 ° from the
vertical. Both the candle-power and the intrinsic brilliancy of
the interior surface are comparatively low at 650 and above.
With the porcelain enameled reflector, conditions are somewhat
different. The reflection is diffuse and the maximum candle-
power from every point on the surface is in a direction normal to
the surface. With diffuse reflection much of the light at each
reflection is directed back into the reflector and strikes the sur-
face again. The consequence is that the entire interior surface
of the reflector has approximately the same degree of brightness.
In looking at such a reflector at angles above 65 ° from the ver-
tical, one sees an interior surface of the same brightness as would
be seen from directly underneath. The candle-power is low at
these high angles because the edge of the reflector cuts off the
light from a large part of the opposite side of the reflector. The
intrinsic brilliancy, however, and the degree of brilliancy contrast
are considerably higher than with an aluminum finish reflector of
the same shape.
These considerations indicate that from the standpoint of eye-
protection, the most desirable reflector is the deep bowl in alu-
minum finish ; second choice is the deep bowl in porcelain enamel
finish, while least desirable from this standpoint are the shallow
bowl and shallow types of reflector.
The question of distribution of light has been covered above
in a general way. It is not necessary here to describe in detail
the characteristics of the extensive, intensive, focusing and dis-
tributing photometric curves. These distributions are well known
and their relative merits and uses have been treated before. It
is important that reflectors should be selected for a given service
by their distribution of light. There is too great a tendency at
present to consider shape of reflector as important in selec-
tion. Having settled the question of eye-protection in any given
installation, shape should be entirely disregarded and distribution
considered. For example, the distributing type of photometric
curve is obtained from two reflectors of entirely different shape,
one in aluminium finish and one in enamel finish. The shape of
these two reflectors should be neglected in choosing between them
and the choice made on the basis of their other characteristics.
6
28o TRANSACTIONS I. Z. S. — PART II
In comparing reflectors on the basis of efficiency, it is im-
portant to consider efficiency in the sense of illumination obtained
for energy expended. The total output of a reflector is no meas-
ure of its illumination efficiency. If this were the case, a bare
lamp would be more efficient that a lamp and reflector. It is
obvious that a shallow reflector will have a greater total output
of light than a corresponding deep reflector, since with the shal-
low reflector much less of the light from the lamp strikes the
reflector and there is, consequently, less absorption at the re-
flecting surface. This greater total output with shallow reflectors
does not mean that a higher percentage of the flux of the lamp
will be useful. Only in exceptional cases is this greater total
output a distinct advantage. In comparing the efficiency of re-
flectors in any given installation, an illumination test is the most
satisfactory method. Certain comparisons from the photometric
curve, however, may be made to advantage when an illumination
test is impractical. The light-flux in the zone o° to 6o°, or
below 6o° from the vertical, is used considerably in comparing
the effective flux of reflectors. This zone includes on the aver-
age, the light striking the plane of illumination directly, in large
rooms. In medium-sized rooms, the angle including the directly
effective flux is somewhat lower. Flux in the zone o° — 6o° is
a much better basis of comparison of the efficiency of reflectors,
than total output or flux below the horizontal.
The specific reflecting efficiencies of aluminum finish and porce-
lain enamel finish are very nearly the same. Porcelain enamel
will usually run somewhat higher. The actual efficiency obtained
from any given reflector, however, depends much upon the shape
of the reflector. With shallow reflectors, porcelain enamel is
considerably more efficient than aluminum. With the deep types,
however, the diffuse character of the reflection from porcelain
enamel tends to cause much of the light to be reflected back and
forth within the reflector. This increases the amount of light
absorbed, and, as a consequence, in the deep bowl reflectors,
aluminum finish is more efficient than porcelain enamel.
The deterioration of reflectors, due to the collection of dust, is
an important consideration in industrial lighting. Dust and
dirt are much more prevalent than in the general lighting field
R0LPH : METAL REFLECTORS FOR INDUSTRIAL LIGHTING 28l
and consequently more apt to reduce the efficiency of the lighting
system. Among metal reflectors the principal difference in this
respect is due to difference in finish. Reflectors with a smooth
interior surface such as porcelain enamel are naturally more
easily cleaned than those with a rough surface such as the usual
types of matte aluminum. Furthermore, porcelain enamel re-
flectors after cleaning regain their initial reflecting efficiency,
while the usual types of aluminum finish, after once becoming
dirty, can never be cleaned sufficiently to regain their original
condition. They always show a permanent deterioration in re-
flecting efficiency of 5 per cent, or more, the exact percentage
depending upon the character of the dirt. It is unfortunate that
this is the case in view of the other advantages possessed by
aluminum finished reflectors. The search for a method of over-
coming this permanent deterioration and difficulty of cleaning
has not been entirely without success. A finish has recently been
developed which will undoubtedly do much to increase the popu-
larity of aluminum. This consists of an interior finish compris-
ing three distinct coats of different material instead of the one
coat of aluminum usually employed. The lower ground coat
consists of a white material which is practically impervious to
moisture. This serves the double purpose of protecting the re-
flector from rust and slightly increases the efficiency since some
light usually penetrates the aluminum finish in places. The sec-
ond coat is the ordinary aluminum finish. The third consists of
a washable lacquer. This forms a smooth hard surface which is
easily cleaned, and when cleaned, restores the surface to its initial
efficiency. This coat is, of course, transparent and the absorp-
tion is so low that the decrease in efficiency due to its use is much
less than the permanent deterioration of the ordinary alumi-
num finish after its first cleaning. In addition to the three inside
coats, the outside of the reflector is finished with the same ground
coat as the inner surface, after which the usual paint enamel is ap-
plied. The reflector, therefore, has 5 coats instead of the 2
ordinarily used. This finish bids fair to overcome the last
objection to the use of aluminum finished reflectors. Aluminum
finish will undoubtedly become more popular than ever before
and gradually replace the less desirable porcelain enamel.
282 TRANSACTIONS I. E. S. — PART II
The comparative cost of various reflectors is a subject which
must be considered for the individual case rather than in any gen-
eral treatment of the subject. Reflectors vary in cost due to dif-
ferences in finish more than due to differences in shape. Of the
finishes in common use, the aluminum is somewhat lower in cost
than porcelain enamel. Initial cost is worthy of very little atten-
tion, however. The value of obtaining the best illumination re-
sults far outweighs such considerations as cost of reflector and
cost of installation. Take for example, the lighting of a single
machine. The best reflector for the purpose may cost 80 cents.
The machine is worth several hundred or very likely, several
thousand dollars. Considering the reflector as a necessary part
of the equipment of the machine it is apparent that its cost is in-
significant. Again, consider the value of the workman's time;
25 cents an hour is a low price, but even at that figure the loss of
only a few hours due to poor eye-protection or poor illumination,
would offset the entire cost of the best reflector obtainable. Or,
take the value of the work performed. An hour's work of a sin-
gle workman may be actually worth several dollars to the com-
pany. Yet one mistake due to poor eye-protection or poor illu-
mination may require many hours' work to rectify it. These
comparisons are sufficient to show that reflectors should not be
selected by cost but rather by the results which they produce.
Illuminating engineering has advanced to such a position that the
best lighting system which a factory can install is usually worth
many times what it actually costs.
From the above treatment of the factors influencing the choice
of reflector, the following generalizations may be made.
1. Aluminum finished reflectors are preferable to porcelain
enamel from the standpoints of
a. Variety of distribution obtainable.
b. Protection of the eyes.
c. Low cost.
2. Porcelain enamel reflectors are preferable to aluminum
from the standpoint of ease of maintenance and lack of
permanent deterioration except when the aluminum finish
is protected by a smooth transparent lacquer; in that case
ROLPH : METAL REFLECTORS FOR INDUSTRIAL LIGHTING 283
the two finishes are probably equally good, for most
classes of service.
3. Deep bowl reflectors are preferable to shallow bowl and
shallow reflectors from the standpoint of,
a. Protection of the eyes.
b. Variety and usefulness of distributions available.
(Deep bowl reflectors are available giving ex-
tensive, intensive and focusing distributions.
The shallower reflectors give only a distrib-
uting type of curve.)
4. Extensive, intensive and focusing photometric curves are
preferable to the distributing curve for the majority of
cases of general lighting. The distributing curve should
be reserved for special cases such as warehouses, stock-
rooms, etc., or work-rooms in which light is required on
high vertical surfaces.
5. Initial cost of reflectors is of minor importance. It should
usually be disregarded, except in comparing reflectors
which are equally good in other respects.
DISCUSSION.
Mr. H. Calvert : It is certainly true that in the industrial
field there are many crimes committed against the conservation
of vision, — bare lamps suspended in the line of vision, improper
reflectors, and lamps badly located. Comparing industrial light-
ing with the lighting of stores, the difference shows up very
much to the disadvantage of the factories and workshops, and
I think the reason is not very hard to find. The storekeeper
naturally wants to attract the public into his store and to do
this he is generally willing to spend a certain amount to improve
his illumination, thinking in doing so that it will increase his
profits. The factory manager, on the other hand, does not come
in contact with the public ; the public does not enter his factories,
and he is very apt to consider that any -money spent on such
illumination means a decrease in his profits.
There is no question that great improvements can be made
by the reasonable use of metallic reflectors. I have in mind a
284 TRANSACTIONS I. % S. — PART II
large weave room which was originally illuminated by means
of 86 series arc lamps with clear globes. These were replaced
by 171 100-watt tungsten lamps with shallow bowl reflectors.
The result was a great improvement in the lighting, greater
uniformity, and a decrease of approximately 40 per cent, in the
energy consumed.
This installation brings to mind an interesting psychological
study. The same mill had a similar weave room which was
lighted practically the same way — the same illumination, but
in the first room of which I spoke the lamps are practically ex-
posed. In the second room a deeper shade is used, so that the
lamps are entirely hidden from view. Now the operators in the
second room are firmly of the opinion that they do not get as
much light as the operators in the first room, simply because
they cannot see the sources of light.
Mr. C. O. Bond: I would wish only to call attention to one
installation of deep bowl metal reflectors, which I think would
appeal to anyone seeing it, that is along the line of the under-
ground platform at the West Philadelphia station of the Penn-
sylvania Railroad Company. As one goes down to take the
train to Baltimore, or going South, there is quite a long platform.
In going up or down such platforms one receives a glare of
light in the eye; and the approaching engineer is so blinded as
he comes down the track that he almost finds himself in the
position of being unable to tell whether he is running into danger
or not. The use of deep bowl reflectors has solved the problem
in this particular case very well indeed and I think it is worth
while for anyone to study that installation.
I should like to ask Mr. Rolph if the local reflection in the
deep bowl reflectors has any influence whatever on the length
of original life of the filament. I was wondering whether there
is a similar effect found by frosting the bulbs of electric lamps,
in which I think the shorter life is somewhat due to the increased
temperature. I ask for that information.
Prof. George A. Hoadley: Mr. Calvert has called attention
to the fact that in many factories there are high ceilings, and I am
ROLPH : METAL REFLECTORS FOR INDUSTRIAL LIGHTING 285
pretty sure that anyone present can appreciate the difficulties in
lighting such places.
Mr. Rolph has spoken of the metal reflectors, and it occurred
to me that in their use we get the type of lamp known as the
direct type. Now is there any particular portion of the light
that comes from these reflectors that is reflected on the walls,
thus giving a bad effect in the illumination of the room ? I have
thought of that as one of the things which might be taken up.
Another point is, that it seems to me it would add very much
to our information if we could have two illustrations — one show-
ing the ordinary daylight, and the other which would give what
we might term artistic illumination. Then we would be able to
make a contrast between the two. But if we have only one, or
only daylight, while we get sufficient information in connec-
tion with the lamp, it does not seem to me that it gives all the
information we ought to get.
Prof. Arthur J. Rowland: Some one ought to write the
history of shades. I presume that on sources of light being
open flames they were originally used for either of two purposes :
to cut light off from a particular direction where it was not
desired, or to serve to reduce the risk of setting fire to
objects in the vicinity. I incline to the opinion that most shades
used on electric lamps, prior to the day of tungsten lamps, were
there because their use was in accord with common practicable
gas burners ; not to produce any special redistribution of light,
or to soften the brightness of a small area light source. Then
again one feels that a bare light is uncouth and crude.
When it comes to shades planned to redistribute light, it is
easy to think there are great differences where none exist. Take
a simple, opaque, cone shade with a white diffusing surface; the
redistribution produced is practically independent of the angle
of the cone or the way it is placed with reference to the lamp ;
except that more or less light is entirely cut off as the light is
pushed further into the cone or partly withdrawn. I am wonder-
ing how much these more modern opaque shades differ from the
simpler and older forms in the light distribution they are able
to produce.
286 TRANSACTIONS I. E. S. — PART II
In Mr. Rolph's paper the word efficiency has been used quite
a little, but nothing has been said about how much the efficiency
is; that is what per cent, of the total light the bare lamp would
pour out, is still poured out from the lamp with any of these
shades over it. It seems curious to me how little manufacturers
care to give that information. Efficiency ought to be as im-
portant as candle-power distribution, for this last has great im-
portance only where a relatively small number of sources is in
use.
I wonder whether some of the first intelligent efforts to pro-
duce efficient shades for ordinary lights giving a desirable light
distribution were not made right here in Philadelphia. Prior to
the time when the names Holophane-D'Olier were coupled as
describing a certain line of opaque shades, a large part of the
line was made by the D'Oliers here in our town. At the begin-
ning of that work, if my recollection serves me correctly Prof.
Edwin Houston, so well known in Philadelphia, made the de-
signs for the shades, the metal used being aluminum. Shortly
after this time the factory superintendent struck trouble. The
aluminum shade with its polished reflecting surface reflected the
light all right but the "streaky" illumination was produced. Some-
thing had to be done. The superintendent had to devise some
remedy. In his difficulty he came out, for some reason or other,
to see me at Drexel Institute. He stated the case and asked me
to suggest some process by the use of which the inside of those
aluminum shades could be given some sort of a matt surface.
I had to tell him I was unable to help, since the question was a
chemical one. I took him to Prof. Henwood, head of the depart-
ment of chemistry who said at once, "Wash them with a little
caustic soda." "Oh," said my friend, "Will anything as simple as
that really do it?" "Yes," said Prof. Henwood, "Nothing more is
necessary — try it." After that they made their whole product
that way.
I had hoped that this evening the discussion would leave the
subject of shades and consider the broader subject of industrial
lighting. There is one very important great detail of industrial
lighting which seems to have been given but little attention by
anyone ; that is, the accumulation of data which will give the an-
ROLPH : METAL REFLECTORS FOR INDUSTRIAL LIGHTING 287
swer to this question, "Is it or is it not worth while to light rooms
and machinery correctly and well ?"
About a week ago I learned from a man who has made an
investigation to determine the effect of artificial lighting on out-
put and on the operator. That nervous strain as indicated by
the presence of nervous headaches, could be eliminated, in the
case of girls working in the printing industries, by using appro-
priately shaded lamps and requiring the operators to wear eye-
shades. He added that the time required on adjusting, making
repairs to machinery, etc., could be very considerably diminished
by a permanent installation of lights at the places where adjust-
ments and repairs were made. The results he said were almost
too good to believe, and that he would not tell anyone how good
they were because he feared the facts might be discredited. He
had not reached the point of considering quality of output, or
safety of employees, when the investigation was discontinued.
Surely some day some one will conduct an elaborate investigation
taking up and determining not only the effect on the output, as
to both quality and quantity, when good artificial lighting is
provided and whether its expense is warranted ; but also how a
worker stands his day's work and how much increased safety is
secured to him by the use of good lighting.
Such questions are as important as any which can be consid-
ered in connection with industrial lighting. The kind of lights
used, their arrangement, the kind of shades put on them, are
insignificant matters compared with the money value of good
light to the industries. This will have to be determined somehow
if industrial lighting is to come into its own.
Mr. J. W. Lee: I should like to ask Mr. Rolph's opinion
regarding glass reflectors and metal reflectors for industrial light-
ing. It appears to me that in large rooms, where the ceilings
and the walls are white, and the lamps are hung high, that glass
reflectors have a particularly pleasing effect — the cheerful aspect
of the light ceilings and light walls have such an effect, to my
mind. ■ Is there any real objection to a glass reflector — which is
more efficient than either the aluminum or metal reflectors —
being used for industrial lighting?
288 TRANSACTIONS I. E. S. — PART II
Mr. T. W. Rolph (In reply) : In regard to the
point that Mr. Bond brought out on the effect of
multiple reflections — I do not believe that I have any-
more information that he has on that point, with the
possible exception of the heating effect. Of course this multiple
reflection is a reflection of heat as well as light, and I can
say that the heating effect has no effect upon the life of
the lamps. Careful temperature tests which have been made
show that the temperature obtained in all types of deep bowl
reflectors — aluminum or porcelain enamel — is not sufficient to
affect the life at all. It is reasonable to suppose that the de-
crease in the effective light of the lamps would be greater for
porcelain enamel deep bowl reflectors, on account of the multiple
reflection back through the bulb; and when the bulb begins to
blacken the absorption will increase. As Mr. Bond pointed out,
it is exactly the same as with frosted lamps. With frosted lamps
the actual life of the lamp is just as great as with the clear
lamps, but the effective light, to 80 per cent, of candle-power is
only about half as great — simply due to the cross-reflection in
the bulb.
Professor Hoadley brought out the point of the reflection
from walls with metal reflectors. Of course with small installa-
tions there is considerable reflection from the walls and the actual
diffusion of course is bad. Diffusion is obtained by light from
distant units in large installations. With metal reflectors there
is not as great diffusion, as a general rule, as with glass reflectors,
but in industrial lighting usually metal reflectors can be so placed
that the direction of the light will be satisfactory for direct light,
so that great diffusion is not necessary.
The question of glass versus steel is one that ought to be
discussed at considerable length if discussed at all. I will
barely touch on it, on the point of efficiency. The most efficient
types of glass reflectors are more efficient than steel when the
ceilings are very light in color. When ceilings are dark in color
the efficient types of steel reflectors give a higher efficiency
than most types of glass reflectors. There is a question of
depreciation to be considered. Of course when bare lamps and
glass reflectors are used to give a higher efficiency in illumination.
ROLPH : METAL REFLECTORS FOR INDUSTRIAL LIGHTING 289
the efficiency which they give is obtained by reflection from
the ceiling to a certain extent, and naturally the depreciation
would be a little greater than with steel reflectors. There are
undoubtedly many cases where glass reflectors are better than
steel reflectors in industrial lighting. But in the great major-
ity of cases it seems that steel would be more satisfactory.
Glass reflectors are appropriate in rooms where the conditions
of dirt are not bad, and in general, in light-colored rooms,
where the work is clean, glass reflectors are quite satisfactory.
As I say, however, that is too big a question to discuss here in
detail. I did not quite get Professor Rowland's point, on the
diffused reflection — on the character of distribution you get with
a reflector when the reflector is used more as a shade.
Prof. A. J. Rowland: No matter what the angle of the re-
flector, or what its size, the redistribution of light is essentially
the same.
Mr. T. W. Rolph : Yes, that is true, with diffusing reflec-
tors ; the general distribution of light from a diffuse reflector
is very nearly the same, no matter what the shape. The shape
of the reflector has some bearing on efficiency — not very much
on the distribution of light — and the extensive curves obtained
with porcelain enamel reflectors are not as good — not as wide,
truly typical extensive curves — as those obtained with aluminum
reflectors.
Professor Rowland asked about the efficiency of reflectors.
Deep bowl aluminum reflectors give a total output of 55 per cent.
to 65 per cent, of the total flux of the lamp. Deep bowl enamel
will run from 50 per cent, to 65 per cent. Some of the deep bowl
in enamel finish will run just as high as in aluminum finish, but
those are reflectors in which the widest part does not come down
quite as far as the corresponding aluminum reflector. The deep
bowl reflectors, which come down well over the filament protect-
ing the eyes well, are a little bit under the aluminum finish in effi-
ciency. In the shallow type of reflector less light strikes the
reflector and consequently shallow reflectors will vary from 70
per cent, to 85 per cent, depending for the efficiency of reflection
upon the character of the material used.
29O TRANSACTIONS I. £. S. — PART II
There is a great deal to be learned, as Prof. Rowland pointed
out, in regard to the effect of good illumination upon the effi-
ciency of the work in industrial lighting. There are many in-
vestigations which ought to be made very much along the same
lines as the investigations which have been made on motion
study and investigations along such lines as that will show
the efficiency of the workers under various systems of illumina-
tion. I have no doubt when we get data of that character it
is simply going to be astounding, in showing the increase in out-
put and decrease in accidents, and such effects of good illumina-
tion. Undoubtedly such investigations will be made in the near
future — a great many of them — because it is too big a commer-
cial question to leave alone very long.
Mr. W. F. LITTLE (Communicated) : Mr. Rolph has given
very clearly his views on the ideal reflector for industrial light-
ing and has put on record his opinions as to the best material
and shape as well as on the proper methods of installations.
He refers also to the very noteworthy development of a more
durable aluminum finished reflector. In the past it has not been
exceptional nor even unusual to find a reflector of this type which
has shown a depreciation in reflecting power of 15 to 20 per
cent, in a short time, even though it had received more than
ordinary care. Once the lustre is gone, an aluminum finished
reflector of the ordinary type has permanently lost much of its
efficiency. If the white protective lacquer referred to by Mr.
Rolph reduces the efficiency but 5 per cent., and prevents further
depreciation, its use marks an important improvement.
Mr. Rolph will probably find many who feel that the flat or
shallow type of porcelain enamel reflector is not such a back
number. In a great deal of industrial lighting where opaque
reflectors are suitable, the conditions are such as to require
general illumination at an angle greater than 65 degrees from the
vertical. Furthermore, by using a distributing reflector of some-
what higher efficiency, throwing some illumination nearer the
horizontal, the contrast between the reflector and background
may perhaps be lessened without a substantial loss in the illumi-
nation on the working plane, thus producing effects more agree-
ROLPH : METAL REFLECTORS FOR INDUSTRIAL LIGHTING 29I
able and equally free from eye strain. Possibly in a majority
of cases this illumination near the horizontal will not be wasted.
It is further implied by Mr. Rolph that for most installations
the bowl type aluminum reflector is preferable to the porcelain
enamel on account of the fact that aluminum produces an
"irregular reflection" while enamel produces a "diffuse reflec-
tion." Also that the light between 65 degrees and horizontal
decreases the ability to see, and that the strain increases with the
angle. Further, that the specific intensity of the reflecting sur-
face will be greater in the enamel bowl type reflector than in
the aluminum reflector of the same type because of the reasons
indicated.
In comparing the two types of reflector it should be noted,
first, that the degree of diffuse reflection secured from porcelain
enamel is largely dependent on the quality of the enamel surface.
It is quite possible to secure enameled reflectors which will con-
trol light almost as accurately as aluminum reflectors. Enamel
of this quality will of course be of a higher order of efficiency.
Second, while it is true that the light between 65 degrees and
the horizontal decreases somewhat, the ability to see, nevertheless
experiments have indicated that the amount of the decrease has
been greatly overestimated and that it is not nearly as important
as some observers have maintained.
Mr. T. W. Rolph (Communicated) : Mr. Little's statement
in regard to enameled surfaces which do not give a high degree
of diffuse reflection is very interesting. If such surfaces can be
used for reflectors and can control the light as accurately as
aluminum finished reflectors, they have a wide field of usefulness.
A reflecting surface, for use on metal reflectors, which will con-
trol light accurately and which can be as easily cleaned as por-
celain enamel and which does not give the disagreeable streaked
effect of polished metal, will be a very valuable addition to the
materials in use at present.
292 TRANSACTIONS I. E. S. — PART II
VISION AS INFLUENCED BY THE BRIGHTNESS OF
SURROUNDINGS.*
BY PERCY W. COBB.
The often-repeated remark that the number of foot-candles
upon the work is not an adequate indication of the virtues of an
illumination system, even solely with respect to that particular
work ; the recounting of experience in corroboration of this fact,
and the experimental attempts that are being made to ascertain
the reason, and give quantitative expression to it, are conditions
which make a very limited amount of introductory matter ade-
quate to the present paper.
This paper constitutes an attempt to answer the question:
"How is vision influenced by the bright, visible surroundings of its
particular object?" The experimental method by which the work
here outlined was done is briefly thus : An observer sitting in a
dark room faces a bright surface of small dimensions trans-
illuminated from the next room. At a given moment the bright
spot thus seen is replaced for a short time by a field of black
and white lines of the same outside dimensions and the same
average brightness. By repeating the experiment with lines of
various widths the exact width of the lines can be determined
which is necessary in order that they may be just visible.
Similarly, instead of using a lined surface, the original blank
surface can be replaced by a field of the same brightness, except
that one half of it is increased or diminished by a small fraction
of its intensity. In this way can be determined the exact differ-
ence necessary in order that it may be seen as a difference. As
both of these quantities (smallest visible detail and smallest vis-
ible difference) vary with variations in brightness, the determina-
tion was made for a series of seven different brightnesses, from
a mere glimmer up to the highest capacity of the apparatus.
These experiments were then repeated with the observer's eye,
* A paper read at the sixth annual convention of the Illuminating Engineering
Society, Niagara Falls, Out., September 16-19, l9^2-
cobb: vision as influenced by surroundings
293
instead of being in the darkness, exposed to white surface illumi-
nated as nearly uniformly as possible. In the work described
this was accomplished by using a large cube — 40 inches (1.01 m.)
side — with its edges and corners filled. The result of this con-
struction was a hollow figure of twenty-six sides each tangent
to a sphere inscribed in the original cube. The interior of this
was painted white and lit by a 100-watt lamp through a milk-
glass which formed one of the oblique sides. The observer placed
0.01 0.1 1.0 10. 42 100
Brlqhtness- eandlerpower per square meter
Fig. 1.— Variations in visual angle (V,) and brightness-difference-perception (1^, M]) with
absolute brightness of test-object, remainder of field is absolutely dark. And same
(Vs, I<s, Ms) surrounding field at a brightness of 42 c-p. per square meter. Abscissae
(brightness) plotted logarithmically, figures giving actual values. Ordinates in the
case ot visual acuity are plotted as visual angle in minutes, subtended by centers of
adjacent dark and bright lines in tesl-object. In the case of brightness-difference
curves the ordinates are actual differences per cent.
his face in an opening in one side and through an opening directly
opposite him he looked at the test-field. The milk-glass through
which the interior was illuminated formed part of the oblique
surface over his head and outside of his 'visual field, so that he
saw nothing but the white, illuminated inner surface of the fig-
ure and the test-field. With the eye under these conditions the
experiments were repeated with known brightnesses of the test-
294 TRANSACTIONS I. E. S. — PART II
field, this latter being kept wholly independent of the illumination
of the interior of the cube. The complete results are given in
Fig. i.
Any one who has ever attempted to determine the limiting
value of a stimulus will know that the exact point of division
between noticeability and the reverse is to be arrived at only as
the properly determined mean of a large number of judgments.
At or near the critical point it is very difficult to be sure whether
one sees lines or not; or whether there is a difference in bright-
ness or not, and if so what its direction is. The observer's verdict
is in any one case obviously to a certain extent at the mercy of
numerous minor influences, and of the state of his own mind.
Further, if it were left to the experimenter's ex tempore judg-
ment just what stimuli to present to the observer the result would
obviously depend to some extent on the experimenter's preposses-
sion in that respect. He would, unconsciously of course, be
influenced to present stimuli at a time when he anticipated an
answer from the observer that coincided with his own anticipa-
tions.
For these reasons, psycho-physicists have devised special meth-
ods for such estimations. A detailed discussion of psycho-physi-
cal methods would be out of place here, but two cardinal points
may be mentioned with a brief account of their application to
the present work.
(i) After determining, by a few preliminary experiments,
approximately how the results will come out, the method of
experimentation is planned so as to be as free as possible from
any arbitrary choice on the part of either the experimenter or the
observer. The whole procedure is cut and dried and no detail
left to judgment in the course of the work if it can possibly be
planned beforehand.
(2) When unavoidable circumstances (such as whether lines
are shown to the observer the first or last half, of an experi-
mental session) would probably influence the result, the indi-
vidual series of observations are so grouped that one half will be
influenced one way, the other half the contrary way, and the dis-
tortion will hence average out. The work has to be so planned,
cobb: vision as influenced by surroundings 295
of course, with respect to every circumstance that may alter the
result.
(1) The single series of observations consisted of sixteen
exposures or stimuli. In the case of lines, for example, the
finest shown in any one series were distinctly too small to be seen
under the test-conditions, the coarsest distinctly visible, and the
fifteen steps between represented a series of equal intervals from
the one to the other. These sixteen stimuli were shown to the
observer in haphazard order, previously determined by lot, and
the observer's judgment rendered as positive, negative, or doubt-
ful, as to visibility of the lines, and recorded.
Similarly, in the case of brightness difference of the two halves
of the field, the two extremes of each series represented distinct
difference in either direction (right half brighter or darker than
left) with equal intervals between.
(2) Each session consisted of four series for each of the two
observers, the other being experimenter at the time. The four
series were : two brightness-difference series, one judged as to
the right side, one as to the left, and two series of visual acuity
judgments. It is in the making up of a session such as this that
factors enter which might have a very serious influence on the
result, i. e., (a) whether observer A came before or after B,
{b) whether he was shown lines before brightness-difference or
after, and (c) whether he judged left or right side first (in the
case of brightness-difference only). The entire set was there-
fore carefully and systematically planned so that A and B came
first each in an equal number of series, lines and brightness-
difference sharing precedence in a similar way, and in the case
of brightness-difference, judgment on the right and left side each
came first in just one half of the series.
The various intensities of the test-field also had to be dis-
tributed each in a fairly representative way over the whole period
of the complete set, in order that any changes in the eye that take
place over the entire period of weeks or -months may take effect
alike on the observations at all the intensities.
The interpretation of the results given in the curves is as
follows :
Visual acuity (V) is plotted as the smallest visual angle in
7
296
TRANSACTIONS I. E- S. — PART II
minutes (from the center of any black line to the center of the
adjacent bright line) under which the lines can be distinguished.
Brightness difference is plotted as (L) the limen, that is the
difference per cent, which, by inference from the results actually
obtained would be correctly recognized in just one half of the
cases in which it was presented to the observer, and (M) the
average difference at which in each series the first deviation from
"correct" judgment occurs.
The results given are the means of all the observations of the
two observers in each case. In spite of the precautions outlined,
the curves in the figure show certain apparently erratic varia-
tions. Whether these are essential or accidental does not appear.
s
>
f\s
,'-°
/
/
•
as
ac
1 0
1 1.
3 IC
100
Briqhtness- candle-power per square meter
Fig. 2. — Visual acuity curves derived from the same values as V! and V2 in fig. 1.
The only way to determine that point would be to multiply the
number of experiments. The results given do, however, show
clearly enough certain features.
The rapid increase in the least noticeable difference at very
low intensities as shown years ago by Koenig, comes out clearly
in the curves Mt and Lx (dark background) as also the low value
for visual acuity (large visual angle) at low intensities (V^.
When the bright background is used the lower limit of vision
is evidently pushed up to a much higher intensity as definitely
shown in M2 and L2, and intimated in V2 by its altered trend.
cobb: vision as influenced by surroundings 297
By far the most striking point in the whole work is, to the
writer's mind, however, the fact that by all three criteria used,
vision at the highest intensity of test object shows a distinct
improvement in the presence of the bright visual field, i. e., vision
is actually improved by filling the visual field with surface almost
as bright as the test-object. The curve showing the visual angle
translated into visual acuity (Fig. 2) brings this out more clearly.
Under certain circumstances then, the eye can see more clearly
when a large amount of light falls into it than when this is cut
off, or in other words, the eye can see equally clearly under those
circumstances with a smaller amount of light on the test-object.
That is, what has been called the "efficiency of the eye" is in
some cases greater with a relatively large amount of light coming
to the eye not from the test-object.
DISCUSSION.
Mr. L. B. Marks : It has been found in practise that an
extremely concentrated lighting field with a dark background
produces visual fatigue. Numerous cases of this kind have come
to my notice in connection with factory and other lighting work.
In considering wall color and brightness of objects within the
field of view, one must take into account the actual intensity of
illumination and the ratio of intensities. In Dr. Cobb's analysis,
for example, if we take very low intensities it might easily follow
that a darker wall would be preferable to a lighter one, while if
we take the higher intensities, the lighter wall would be preferable.
Mr. J. R. Cravath : The point which interests me especially
in Dr. Cobb's paper is that of the value of illumination on the
background ; namely, 42 candle-power per square meter, is a value
which is frequently attained in practise on light colored walls.
We would have good reason to suppose, with the information
which the paper gives, that we would have higher visual acuity
under practical working conditions with such rather highly illumi-
nated walls. Dr. Cobb does not attempt to explain why this is
so. It has been suggested that the diameter of the pupil is less
where we have bright surroundings and therefore we are able to
focus more clearly, just as we get a clearer image in the camera
with a small aperture.
298 TRANSACTIONS I. E. S. — PART II
Dr. H. E. Ives: I have a little fact to bring forward in line
with Dr. Cobb's research, in connection with the photometry of
lights of different colors. You may have noticed that in my
paper this morning I recommended that the photometric field be
surrounded by a larger field of equal brightness. I did not go
into details in my abstract this morning in regard to that, but I
might enlarge upon it now.
In comparing lights of different colors by the equality of bright-
ness method, I found by experiment that the presence of bright
surroundings to the photometric field resulted in the range of
settings being approximately divided by two ; in other words, the
sensibility was enormously increased.
As regards the question of fatigue, it has been my experience
that the eye becomes very much fatigued looking down the black
tube of the ordinary optical instrument at a bright field ; on the
other hand, using a bright tube, one could work very much longer.
There was really a tremendous difference. In the paper this
morning this observation was incorporated and the reason given
is that it makes for the comfort of the observer. I am giving
this for what it is worth. My own interpretation of it is that
bright walls are not quite as black as they have been painted.
I do not think any difference in the diameter of the pupil would
explain the differences found in the discrimination of shade differ-
ences, which show the same characteristics as the visual acuity
tests.
Mr. H. P. Gage: I would just like to say a word about the
last part of the paper, where Dr. Cobb shows that it is possible
to detect differences in brightness and fineness of detail with a
brighter surrounding field than with a dark surrounding field.
I think this is due to the fact that the eye is able to focus accu-
rately on the test object, whereas with a dark surrounding field,
the eye is out of focus and the object is not seen clearly.
Another point, a little outside of this, was brought out by the
use of the acuity method, showing that a monochromatic light
will give a higher acuity than white light. This has been pretty
well recognized as due to the chromatic aberration of the eye.
It would be very interesting to try some of these tests, using a
glass in front of the eye designed to make the eye an achromatic
instrument. The normal eye at rest will be found to be in focus
VISION AS INFLUENCED BY SURROUNDINGS 299
for the red end of the spectrum, if the eye is correctly focused
for a red light and it will be found that for blue or green light,
the eye will be short-sighted. This has an important bearing on
signalling.
The other night I had a negative lens and I took occasion to
examine the lights from the rear of a train (red, green and
yellow). It was interesting to note the greater range at which
the red light could be seen under the ordinary conditions of the
eye at rest. With the negative lens, which renders the eye far-
sighted, the green light, which appeared as a bright point with
rays around it, was brought down to a definite round spot and
could be seen at a much greater distance than without the glass.
This indicates the reason why different people, not color blind,
will get different appearances with the same colored signals.
In signal work, not only the question of color sensitiveness and
color blindness enters into consideration, but also the dioptric con-
dition of the eye whether near or far-sighted and, if normal, for
which particular wave-length the eye is in focus when at rest.
Mr. M. Luckiesh : The data given show that under the con-
ditions of the test, better results are obtained amid bright sur-
roundings when the brightness of the object is 10 to 100 candle-
power per square meter. The bright surroundings were not any
brighter than those very often found in actual practise.
In regard to fatigue : the periods of the test were about 45
minutes long for the observer. This is not considered by psy-
chologists to be long enough to induce undue fatigue. In fact
Dr. Cobb's data do not show any effects of fatigue when the data
of the first and last halves of the period are compared.
The question of better acuity in monochromatic light was quite
thoroughly thrashed out at the convention of the Society last
year.
Dr. P. W. Cobb (in reply) : Referring to Fig. 2 — at the level
where visual acuity is 1.5 the bright surroundings evidently from
the (dotted) curve gave the eye an advantage, since vision main-
tained this value with a less bright test object than was necessary
in dark surroundings. That is, the "efficiency of the eye" was
greater than unity — in fact (bearing in mind that we are using a
logarithmic abscissa-scale) the "efficiency of the eye" estimated
300 TRANSACTIONS I. E. S. — PART II
from the curves would be here about 2.00. At 5 candle-power
per square meter (about one-eighth of the surrounding bright-
ness) the two curves cross — which means that this value is here
unity, and below this point it is less than unity. It could evidently
be made almost indefinitely small by selecting a low enough test-
object brightness to start with. The curves show it to be about
0.15 at the lowest point given (visual acuity a trifle over 1.0).
I would add that the visual acuity values came out much more
consistently and definitely than the brightness-difference values
and are hence much more trustworthy. However, as the curves
show, the change in conditions works a relative change in results
in every case of the same direction as that seen in the case of
visual acuity.
I admit that a more thorough investigation of the matter in
the region of higher brightness is desirable as far as the M and L,
curves are concerned. That is easier seen than foreseen. Each
point in the figure indicated by a small circle is the result of 512
(in a few cases only 448) separate judgments, each pronounced
after 3 seconds observation of the test object. The entire work
of which these are the final results extended over the larger part
of a year.
In reply to Dr. Gage, I would say that the surrounding surface
was 1 meter distant, while the test-object itself was 2 meters
from the eye. In view of this, accommodation for the test-object
should be more disturbed when this latter was lit up. There was
no evidence whatever of any difficulty of this kind. The blank
field was always present to the eye before the exposure, having
borders identical with those of the test-field which served as a
guide to the eye in accommodation.
The observation cited by another speaker is interesting. The
conditions in the case he mentions (light from a source directly
upon the eye) are not the same as an equal number of meter-
candles coming from an extended and uniformly bright surface
to the eye. As a matter of fact, working with the brightest test-
object on a dark field produced considerable discomfort and
vaguely uncertain feelings referable to the eyes, which did not
appear in the presence of the bright surroundings.
It is exactly such ideas as those that Mr. Marks has just
expressed that have made this work appear to me to be worth
VISION AS INFLUENCED BY SURROUNDINGS 301
while. Nature, rarely if ever, treats man to an extreme local type
of illumination. We must bear in mind that sunlight is about
one thousand times as intense as ample artificial illumination ; so
I do not feel that the brightness chosen for the background in
this work is at all abnormal to the eye, although it may not fall
within the limits of engineering practise.
Replying to Mr. Cravath : I have preferred in this paper to
give the results without going into the theory of the matter. A
smaller pupil does not necessarily mean more distinct vision
except where the refraction of the eye is imperfect. In my own
eye (moderate astigmatism, corrected) I found by using a set
of artificial pupils that visual acuity grew markedly less with
decrease in pupillary size below 3 mm., even when the brightness
of the test-object was raised to compensate exactly the decrease
in pupillary area. And further, the M and L curves, Fig. 1,
show changes in the same direction that visual acuity does, and
it is fair to suppose that exactness of the retinal image plays at
most a much less significant part in the estimation of brightness
differences in visually gross areas. I do not wish to be under-
stood, however, as saying that the pupillary diameter has been
eliminated as a factor. I do feel that other things are necessary
for a full explanation.
There is in progress at the laboratory with which I am con-
nected further work of similar character to this, using a bright-
ness of 3 candles per square meter for the surroundings instead
of 42, which it is hoped will make the entire question more clear.
302 TRANSACTIONS I. E. S. — PART II
A PRACTICAL SOLUTION OF THE PROBLEM OF
HETEROCHROMATIC PHOTOMETRY
BY PROF. CH. FABRY.
Synopsis: In order to reduce all practical photometry to that of lights
of the same color, secondary standards of various colors or colored
absorbing media are necessary. These should be calibrated in the
standardizing laboratory, by the most refined methods of heterochromatic
photometry. The calibration of an infinity of colored standards, or of
every individual colored glass, is not practical. Herewith are described
absorbing liquids, definitely specifiable, which may be used in varying
thicknesses and proportions to make the light of a given standard like
that of any other illuminant. A yellow and a blue solution have been
found which suffice to match with a Carcel, all the ordinary illuminants;
a purple solution is suggested for use where these are not sufficient, the
three absorptions giving, by the three-color principle, all the tints as
represented in a color triangle. The method of use of these screens is
outlined, the possible methods of calibration are described and some
experimental results are tabulated.
In spite of many good works, the problem of heterochromatic
photometry seems not to be satisfactorily solved for practical use.
In the comparison of two lights of very different colors, as, for
instance, an incandescent carbon lamp, used as a standard, with
daylight or the mercury-vapor arc, there remains a large amount
of uncertainty and arbitrariness. Several very good papers,
published in these Transactions, especially by Dr. Ives, have
forwarded the question. Probably the definitive solution will
be conformable to the suggestion of Dr. Ives1 namely : Measure
every simple radiation in quantity, by the energy conveyed by
it in one second, an operation entirely independent of the prop-
erties of the human eye; then, quote every one at its proper
value for the special application to lighting, this value being zero
for all infra-red or ultra-violet radiation. Thus, for computing
the value of some ore containing metals of different prices (gold.
silver, copper, etc.) one will determine the weight of each metal,
and each one will be quoted at its proper price, sand and stone
being quoted at zero for metallurgic purposes.
For this definitive solution we are not yet ready, at least for
1 Transactions of the Illuminating Engineering Society April, 1911, page 258, and
Astrophysical Journal. November, 1912, page 322.
Fabry: problem of heterochromatic photometry 303
everyday use. Meanwhile, we are obliged to choose some pro-
visional solution, and a French proverb says that nothing lasts
longer than a makeshift. It is moreover possible that a pro-
visional solution could be turned into a definitive one when it
will be possible to properly express the experimental results.
In actual practise, we cannot think of making a complete analy-
sis, qualitative and quantitative of every light to be measured.
So long as the eye will be used in photometric measurements, the
equalization of two illuminations on the photometric screen is
all that can be asked for the ordinary observer, and this equaliza-
tion can be accurately made in only one case, i. e., when the lights
are of the same color. We are thus led to eliminate from prac-
tise any heterochromatic measurement, as was excellently ex-
pressed by Dr. Ives :2 "I think it extremely important — it is
essential — that in ordinary photometry there should never be
made a comparison of different colors. All practical photometry
should be reduced to photometry of the same color. Conse-
quently, the question of which photometer is to be used for
comparing lights of different colors, becomes a question for the
standardizing laboratory, the Bureau of Standards, the Reich-
sanstalt or the National Physical Laboratory."
What have we to do in order to reach this end, viz., to have
only comparisons between lights of the same color? We cannot
hope to find as many different standards as there are different
tints, i. e., an infinity : the difficulties encountered in finding one
good standard do not encourage one to search so large a number.
With a single photometric standard, the only chance of suc-
cess is to modify its tint by interposition of some apparatus in
the beam of light, in order to give it the desired color. This
change of color will at the same time modify the intensity in a
ratio which will be measured once for all in every case ; this
measurement, unavoidably involving heterochromatic measure-
ment, will be made by the standardizing bureau.
There are many means at the hand of, physicists by which the
tint of a complex radiation, i. e., the proportion of the different
simple radiations in the light under consideration, may be
2 Transactions of the Illuminating Engineering Society, November, 1910, page 727.
304 TRANSACTIONS I. E. S. — PART II
changed at will.3 From a practical standpoint, the simplest
means is the interposition of some absorbing medium. The use
of such absorbing substances (colored glasses) has been sug-
gested and actually tried.4 The procedure is as follows : We
have, for instance, to compare an electric arc with an incandes-
cent lamp used as a standard ; when the two parts of the photo-
metric screen are illuminated by the two sources, they appear
very different in color, and the equalization of brightness is
largely a matter of judgment. On the incandescent lamp a blue
glass properly chosen is superposed, which absorbs the excess
of red and yellow light contained in the radiation of this lamp,
and gives it the same color as that of the electric arc. The
equalization of illumination on the photometric screen will then
become easy and precise. By this procedure, we are able to com-
pute the intensity of the arc lamp if we know how many candles
the incandescent lamp gives through the blue glass. As we
know the intensity of the lamp without the absorbing glass, we
need only to have the ratio between incident light and trans-
mitted light through the glass, i. e., the opacity of this glass for
the total radiation of this lamp. The opacity can be measured
once for all, but this value is right only for the glass employed
and for a lamp giving the same kind of radiation as the lamp
used for the measure. If a higher voltage is applied to the lamp,
the proportion of blue radiations freely transmitted through the
glass will be greater, and the opacity will be less. The measure
of these opacities for fixed light would be the problem of the
standardizing Bureau, and as it is probably impossible to repro-
duce glasses exactly, it would be necessary to make the measure-
ment on every separate sample of glass, of which a great many
would be necessary to match the color of the different sources
now employed in practise.
An important step would be reached if it were possible to use
3 All these means work by inequal weakening of the different simple radiations;
they are incapable of adding simple radiations lacking in the considered light. If this
was monochromatic, any one of these means could not change its composition. We con-
sider, in fact, as standard source, such a source as giving forth a mixture of every simple
luminous radiation. In such a case, it is conceivable that, by proper weakening of the
different simple radiations, it would be possible to get every possible mixture of simple
radiations, i. e., every possible light.
4 Ives, Transactions of the Illuminating Engineering Society, November. 1910, page
728; Cady, October, 1912, page 385.
FABRY : PROBLEM OF HETEROCHROMATIC PHOTOMETRY 305
definite absorbing media, reproducible by every observer; the
opacities of different thicknesses might be measured once for all,
and every observer would be able to use these values, without
purchasing a definite object measured at the bureau. The light
used for these measures of opacity should be of definite com-
position, but it need not be the light of a standard of intensity ;
it is possible, and perhaps better, to separate the two duties, in
order to choose the best standard for each purpose ; the standard
of intensity ought to have a fixed intensity, and a very small
change of tint from one standard to another is of no conse-
quence ; the standard of color must give a radiation of constant
composition, but a change of intensity from day to day is abso-
lutely unimportant because it will every time be compared with
the standard of intensity.
These thoughts came in my mind ten years ago, and I made a
long series of experiments to put them into practise. Some
physicists to whom I showed my results raised many objections,
from which I concluded that the theory of colors was not yet
sufficiently understood to permit the general use of my method ;
I published only a short article,3 which was entirely unnoticed.
The situation is now changed,6 and I think it is now worth while
to publish a more complete account of my experiments, in the
hope of contributing to the improvement of the methods of
heterochromatic photometry, which are still in a somewhat chaotic
condition.
I suppose the question of the standard of intensity to be solved.
In the practise of to-day, the standard is not a flame, but an
incandescent carbon lamp which, at fixed voltage, gives a definite
number of candles.
Furthermore, I chose a source giving light of definite tint, with-
out respect to intensity, which need only remain constant during
the short time of the measure. I have chosen the old Carcel
6 Comptes-Rendus de V Aeademie des Sciences, 9 Novembre, 1903, (Sur une solution
pratique du probleme de la photomfitrie heterochrome).
* If, today, the technical men are better prepared to understand the problems of
heterochromatic photometry, that is in a large degree owing to the works published by
the Illuminating Engineering Society ; it is for that reason that I have wished my paper
to be published in these Transactions, the readers of which are the best prepared to
grasp the problem here dealt with.
306 TRANSACTIONS I. E. S. — PART II
lamp,7 which is probably good enough for this very simple ser-
vice; perhaps a carbon incandescent lamp at a fixed number of
watts per candle would be more convenient. The light of the
Carcel is very similar to that of the carbon lamp ordinarily used
as standard of intensity, but is very different from many other
lights, especially from the light of the sun.
I have sought such colored substances as, interposed in the
light of the Carcel, will give it the color desired in every case.
To have reproducible absorbing media, whose opacity can be
measured once for all, one is lead to use liquids of definite com-
position and proper thicknesses. It is desirable to be able to get
every possible tint, in order to match the color of the standard
with that of each light. A single liquid is not sufficient to arrive
at this end; if only one liquid is used, only one variable is at
one's disposal, viz., the thickness of this substance or, what is
the same, the concentration of the absorbing dye in mixture with
an uncolored liquid as water. A single set of colors can be thus
obtained ; if every tint is represented by a point in the triangular
diagram of Maxwell-Koenig,s the different tints produced in vary-
ing the thickness will be represented by the different points on
the curve A (Fig. i) beginning at the point M which represents
the tint of the Carcel without the absorbing medium (thickness
zero). Only the colors represented by some point on the curve A
can be matched by the use of the Carcel with the absorbing
medium under consideration.
Another liquid of different color will give another set of tints
7 I did not use the Hefner lamp, on account of its too red color, and because it is too
weak when further weakened by absorption. The light of the Carcel is so similar to that
of the carbon incandescent lamp at normal voltage that the values given below would
probably be applicable without change to this latter source.
B For the definition of this diagram, see : Ives, Transactions of the Illuminating
Engineering Society; April, 1910, page 205, and April, 1911, page 266.
FABRY : PROBLEM OF HETEROCHROMATIC PHOTOMETRY 307
represented on the diagram by the different points on another
curve B beginning at the same point M.
If two absorbing cells are used, one of the first liquid and one
of the second, we have at will the values of two variables, the
thicknesses of the two cells ; it will be possible to obtain all the
tints represented in a certain part of the diagram. We cannot
cover all the surface of the diagram, but, at least, a finite surface
of it. If the two liquids are properly chosen, we can hope to
match exactly the light of many of our ordinary sources.
The two liquids chosen are one of blue color, with absorption
increasing from violet to red, and one yellow with absorption
varying in the inverse direction. The aim to have liquids definite
1 i
1 1
1 1
Fig. 2.
and reproducible leads us to set aside all the anilin dyes, and to
seek only among the mineral substances. The two liquids chosen
were :
A. Crystallized copper sulphate 1 gram
Commercial ammonia (density 0.92) 100 cubic centimeters
Water, quantity sufficient to make 1 liter
B. Potassium iodide 3 grams
Iodine 1 gram
Water, quantity sufficient to make 1 liter
The thicknesses of the cells containing each of these liquids can
be varied at will ; I shall denote by x the thickness of the blue
liquid A and by y the thickness of the yellow liquid B, the two
expressed in millimeters. The thickness can be varied at will if,
for each liquid, a double wedge shaped cell is used (Fig. 2) ;
when one of the wedges is slid, the totar thickness of the liquid
can be varied continuously. It is ordinarily simpler to leave
constant the thickness, using a parallel cell, and change the con-
centration of the liquid ; I have verified the fact that a change in
308 TRANSACTIONS I. E. S. — PART II
concentration produces the same effect as a change of thickness
(Beer's law) ; x and y are then the products of the thickness
measured in millimeters by the number of grams per liter (copper
•ulphate or iodine) in the solution used. In this change of con-
centration, one must leave approximately unchanged the propor-
tion of ammonia contained in the liquid A, and consequently add
to this liquid not pure water but water containing ammonia in
the proportion of ioo cubic centimeters to the liter. For the
liquid B, pure water can be added. I ordinarily use cells with
parallel faces and of 20 millimeters thickness.
With these two absorbing media of proper thickness and con-
centration, and the Carcel lamp used as standard of color, it will
be possible to match the color of almost every source used in
practise. With a single cell containing the liquid A (y = o) we
obtain, with increasing thickness (x greater and greater) light
of color more and more blue ; we can match in this way the color
of every source similar to the black body at higher temperature
than the Carcel, including the solar light. The tints produced
in this way are represented in the diagram (Fig. 1) by the differ-
ent points on the curve A. With a single cell filled with the
liquid B, by increasing the thickness or concentration of this
liquid (x = o, y increasing), we obtain lights of more and more
yellow and finally a red tint; the colors so obtained are repre-
sented on the diagram by the curve B. Using the two cells with
every possible thickness or concentration (x and y varying at
will), we can produce light represented to the left and above the
curve AMB, viz., all the set of blue, green, yellow or red tints,
more or less saturated. We cannot get the tints represented to
the right and under AMB, i. e., the purple tints, in which are
predominant the radiations of the two ends of the spectrum with
weakening of the middle part ; a third liquid, used with one or
the other of the two just described, would be necessary in order
to get these tints. I did not think it worth while to seek such a
liquid.
For every source of light we have to search for the values of
x and y necessary to match its color. The following table gives
some values.9
9 These values are given here only as an indication, inasmuch as the indicated sources
are by no means in themselves definite in color.
FABRY : PROBLEM OF HETEROCHROMATIC PHOTOMETRY 309
x y
(Blue (Yellow
Source of light solution) solution)
Ordinary carbon arc (not mineralized) 33 0.0
Tungsten lamp 10 0.0
Acetylene flame 12 0.0
Incandescent carbon lamp at */j of its normal
voltage o 5.4
Id. at :| , of its normal voltage 19 0.0
Hefner lamp o 1.5
Auer light 41 1.0
Nernst lamp 22 0.6
Cooper-Hewitt lamp 76 0.6
Sun at noon in summer 54 0.0
As can be seen, the blue solution is, in most cases, the most
important, and indeed in many instances, must be used alone to
produce the desired tint. It is the case for all sources of light
whose spectrum is similar to that of a black body at higher tem-
perature than the Carcel (tungsten lamp, electric arc, acetylene,
sun). The precision with which these sources of light can be
matched by means of the Carcel and copper liquid is really sur-
prising. The tint of the sunlight, so greatly different from that
of the Carcel as to render almost photometric comparison im-
possible, is practically indistinguishable from the color of the
modified Carcel.10 That does not prove that the energy curves
of these two lights are entirely identical, but the energy curve
of the so modified Carcel is much more like that of the sun than
is the energy curve of the natural Carcel, as is to be seen from
Fig. 3 where are represented the energy curves of the Carcel lamp,
of the same source through 54 millimeters of copper solution and
that of sunlight. For a normal eye the two last tints are identical,
and it is probably so for every eye not entirely abnormal. The
white stars (like Vega) give a light still more blue than that of
the sun ; they could probably be matched with the Carcel light
by using a somewhat greater thickness of the same liquid. For
the explanation of this property of the copper solution, see later
(appendix).
The interposition of colored cells in the radiation of the Carcel
'" I have many times used the light obtained in this way for works of astronomical
photometry (determination of the candle-power of the Sun, comparison of the light of
the Sun with that of the Stars, etc.).
3io
TRANSACTIONS I. E. S. — PART II
lamp does not only change its color ; the intensity is also weak-
ened. We must know this change in order to use our absorbing
media in photometry. The weakening is partly produced by
reflection on the faces of the cells, and partly (the greatest part)
by absorption in the liquids. In order to eliminate the effect of
reflection, we shall compare the intensity of the Carcel through
cells filled with pure water with the intensity through the same
cells containing the liquids ; the ratio of the first number to the
second will be called the opacity of the liquids. This opacity is
a perfectly definite function of the two quantities x and y which
characterize the conditions of the two media. It is possible to
investigate it once for all, and get a numerical table or an empi-
rical formula giving the values of the opacities for every value
&5u o.6u
'WAVE- LENGTHS '
Fig. 3-
of x and y. The experiment necessary to get the numerical
values of the opacity are, unavoidably, heterochromatic photo-
metric measurements ; they ought to be made with the most
refined methods of heterochromatic photometry and, if possible,
by several observers, to have the most probable values for a
normal observer.
I made some such measurements in 1903; they were not pub-
lished. I will give here my results, but I do not claim for them
great accuracy : the method used was that of equal brightness,
with the use of the Lummer-Brodhun photometer (the flicker
photometer was not yet in general use) ; perhaps the brightness
of the photometric screen was not great enough to eliminate the
Fabry: problem of heterochromatic photometry 311
complications arising from the Purkinje effect, which leads to
over-estimates of the intensities of blue lights ; the measurements
were made only by one observer. I have some reason to think
that the values given for opacity of the blue solution were rather
too low. Be that as it may, I have found that the opacity for
thickness x and y of the two luiquids is expressed by the follow-
ing empirical formula :"
log opacity = 0.016 x -f 0.032 y — 3 X io-5 x- + ( 1 )
6 X 10-4 xy
The following table gives some values computed from this
equation :
Opacity
Opacity
0
0
I. OO
0
1
I.08
10
0
1.44
0
2
1. 16
20
0
2.02
0
4
1-34
30
0
2.97
0
6
1.56
40
0
4.0
10
1
1-57
50
0
5-4
20
2
2.49
60
0
7-3
40
4
6-55
60
6
18.2
The values of the opacity being known, it becomes easy to
reduce every photometric measurement to comparisons of the
same color. Suppose we have, for instance, to measure the
intensity of an electric arc in terms of an incandescent carbon
lamp used as standard of intensity. We will first seek what
thickness (or concentration) of the two liquids must be inter-
posed in front of the Carcel to give to its light the color of the
electric arc; we will find that we must have no yellow cell, and,
for the blue one, a thickness or concentration defined by x = 33.
On the other hand, the incandescent lamp has almost exactly
the same tint as the Carcel ; we need not change the color to
compare these two lights. On the photometric bench we put
on one side the standard lamp, on the other side the Carcel,
interposing the cell to be used later, filled with water ; the equal-
ization of the two intensities will be made. Instead of the
standard lamp we put then the arc to be measured, and we fill
the cell placed in front of the Carcel with the proper blue liquid ;
we equalize once more the illuminations, which are still of the
11 For the derivation of this formula from a theoretical standpoint see appendix.
8
312 TRANSACTIONS I. E. S. — PART II
same tint. The opacity of the used liquid being known, the
computation of the intensity of the arc is easy.
This procedure is scarcely more complicated than the ordinary
measurement with lights of the same color; it is true that we
must equalize twice the illuminations on the photometric screen,
but it is ever so when the double weight method is used and
the use of this method is generally to be recommended. If we
have many measurements to be made with the same kind of light,
the same liquid will be used. Perhaps, in this case, it will be
found more convenient to use some colored glass properly chosen,
if such a glass can be found; it will be easy to measure the opac-
ity of this glass by comparison with the known liquids, without
going to the standardizing bureau, and by making comparisons
of nothing but lights of the same color.
It is worth remarking that the standard color lamp (Carcel
for instance) with the absorbing liquids gives us an arbitrary
scale of tints whose use is very convenient and quite inexpensive.
True, this scale is quite arbitrary, but from the data so obtained
for a light, it is always possible to pass to the tint expressed in
absolute value, i. e., the values of fundamental sensations con-
tained in the studied light, or the point which represents it on
Maxwell's diagram. In some cases where the other methods are
difficult to use, perhaps this procedure could give some useful
information ; it would be perhaps so in some astronomical ob-
servations.
As was said, the use of two liquids is not sufficient to produce
every possible tint; with the two liquids chosen, we can only get
the tints represented above the curve AMB (Fig. i). With a
third liquid we could get every possible tint. The liquid to be
chosen ought to be purple, absorbing the middle part of the'
spectrum more than the two ends; this medium, taken alone
with increasing thickness would give tints represented by such
a curve as MC. On combining this liquid with the blue liquid
A, it would be possible to have every tint represented by points
in the angle AMC ; the same liquid in combination with B would
allow us to cover the angle BMC. In short, we should have
three liquids giving respectively, when employed separately, the
curves MA, MB, MC; combinations of two liquids allow the
Fabry: problem of heterochromatic photometry 313
production of every possible tint. Perhaps a solution of potas-
sium permanganate would be good as a third liquid.12
The absorption methods in optics have been often criticized,
and sometimes rightly, because many incorrect results have been
noticed on account of an incorrect interpretation of the facts.
Certainly, the use of absorption, which does not allow the com-
plete separation of simple radiations, demands a critical attitude
on the part of the inventor; but, when the methods are properly
tried out, this process leads to very simple experimental devices,
which is of first importance from a practical standpoint.
APPENDIX.
I will collect here some calculations and numerical data not
necessary for the practical use of my method, but useful for its
comprehension.
I have investigated the absorption curve (absorption as a func-
tion of the wave-length) of my liquids. The measurements were
made by use of a spectrophotometer, without seeking for a great
accuracy, but in order to find the behavior of the phenomenon.
Let us consider -the liquid A ( 1 gram crystallized copper sul-
phate and 100 cubic centimeters ammonia in 1 liter), and take
it with a thickness of 1 millimeter (.r = 1). For a simple radia-
tion of wave-length a, this cell has an opacity w (ratio of the
intensity transmitted through water and intensit3r through the
cell). Instead of w it is often more convenient to characterize
the absorption by the decimal logarithm ot w, called the absorp-
tion constant a:
a = log opacity = absorption constant.
If the cell has a thickness x the value of the absorption constant
for the same wave-length will become a.r.
For the liquid B ( 1 gram iodine and 3 grams potassium iodide
in a liter) in thickness of 1 millimeter (y = 1) the similar quan-
tity will be denoted /?.
Fig. 4 gives the values of a and ft plotted against the wave-
length.
If we have to investigate the action of- these absorbing media
on the light of the Carcel, we must know the luminosity curve
12 The use of the third liquid would probably be necessary to match the color of certain
flaming arc. I did not make experiments on these sources.
3H
TRANSACTIONS I. £. S. — PART II
of the spectrum of its light. We can admit that the radiation
of the Carcel lamp is identical with that of a black body at a
temperature a little under 2,000 degrees of the absolute scale;
its light is not so yellow as that of the Carcel, but not so blue as
that of the acetylene flame. We can therefore trace the curve
which gives, as a function of the wave-length A, the intensity
I measured by the quantity of energy carried in one second
(energy curve of the Carcel). On the other hand, we know the
sensibility curve of the eye as a function of the wave-length13 ;
let us denote by 8 the sensibility of the eye for the radiation A;
0.6
2
O
|0.5
0
§0.4
I0-3
0
0
"0.2
0.1
0.024
0.022
0.020§
F
o.oia:>
0.0 1&°
111
0.014 ji_
0.0 12 t
o
0.010%
o
0.0083
0.006 ^
0.004
0.002
7r
Fig. 4.
the values of S are proportional to the luminous intensity at equal
energy value. The product L = 18 is the luminous intensity (in
arbitrary units) for the radiation A, and we can trace the
luminosity curve of the spectrum of the Carcel.
We put in front of the Carcel the thickness x of the blue liquid;
the opacity of this cell for the radiation A is io"^; this radiation,
whose luminous intensity was L, will be reduced to I, X io— ax.
It will be easy to trace the luminosity curve of the light so modi-
fied.
In order to have the opacity of the cell for the complete light
of the Carcel, we must compare the total intensities without and
with absorption. In the first case, the total intensity is :
13 See, for instance : Ives, Transactions of Illuminating Engineering Society,
April, 1911, page 261, and October, 1912, page 379; or Astrophysical Journal, November, 1912
Fabry: problem of heterochromatic photometry 315
With absorbtion, it is:
fLio-^A..
The opacity ft, ratio of the first number to the second, is:
a
jW
rdk
A similar calulation will be applied to the liquid B.
If we use the two liquids A and B, we will find for the opacity:
0 = 7 (2)
jLicH^ + fi-v)dK
As the values of L, a and /? are known for every value of A, we
can compute the values of the opacity for every value of x and y.
The integrals are computed graphically, and that is not difficult
if one uses a planimeter.
I have made this numerical computation for some values of
x and y. The concordance with experimental values given above
is not perfect, but tolerably good, in consideration of the lack of
accuracy of the numerical data used (luminosity curve of the
Carcel, absorption curve of the liquids) and of the uncertainty
of my heterochromatic measurements.
The equation 2 which gives the opacity as a function of x and y
is not expressible by the elementary functions. We can express
it by a development in series. The mathematical study of the
question shows that the best way is to develop log O in increas-
ing powers of x and y. Dropping powers of the variables higher
as the second, we find :
log ft = Ax + By + C.v2 + D/ + Uxy.
The values of the co-efficients are:
316
TRANSACTIONS I. E. S. — PART II
A =
J-
hdX
C = 1,151
D= 1,151
3.302
B f
\I<d\ ^d\
~(\phd\\ j/3'WA.
(Jttft)' jwx
It can be proved that A, B and E are positive, while C and D
are negative. The formula is of the same type as the empirical
equation 1 ; the computation of the coefficients by the above
formula gives values approximately in accordance with those
empirically obtained.
Lastly, we can compute the color produced by a given thick-
ness of the two liquids interposed on the Carcel, and represent
every tint by a point on Maxwell's diagram. In this connection,
I have above insisted on the precision with which the blue solu-
tion alone, interposed in front of the Carcel, can change its color
in such a way as to give it the color of the radiation of any
black body at higher temperature. To explain this property, we
will examine the following problem :
The Carcel is supposed to have the same radiation as a black
body at a temperature T (a little under 2,000 degrees). We wish,
by a properly chosen absorbant, to modify its radiation in such
a way as to give it the same composition14 as that of a black body
at a temperature T' higher than T. What absorption curve must
this absorbent have, i. e., what is the relation between the absorp-
tion constant of the cell used and the wave-length?
14 That is a sufficient but not necessary condition to have the desired color. If it is ful-
filled, the tint will be the same for every eye, even not normal.
Fabry: problem of heterochromatic photometry 317
The problem is immediately solved if we know the energy
curve of the two radiations, and these curves are denned by
Wien's formula.15 The opacity of the cell ought to be the ratio
of the two intensities, multiplied by an arbitrary constant. We
find thus that the absorption constant a (logarithm of opacity)
must be expressed as a function of the wave-length by the equa-
tion :
« = 6,3.o(i-4;)!+C'<=-x+B.
A and B being two constants.
If we trace the curve giving the values of a as a function of
- we must have a straight line. If such a substance is found.
A
it will allow us to transform the radiation of a black body at
temperature T into a radiation identical with that of a black
body at higher temperature T1 ; the thickness of absorbant to be
used is proportional to -7=- — =7 .
Returning to our blue liquid, if we trace the curve giving its
absorption constant as a function of — , we have the curve in
A
Fig. 5. That is not mathematically a straight line, but the curve
has, in the brightest part of the spectrum, a very long inflection ;
the part MN, the most important for color, is almost linear.
It is, therefore, not surprising that the color (not requiring abso-
lute identity of energy curve) of the black body at any high
temperature could be got by the use of our liquid. In fact, the
thickness necessary to make the transformation is approximately
given by the equation :
16,500 \y~ "TV •
The determination of the value of x necessary to match the
color of the Carcel with that of a source of light gives us an
indication of the temperature of this source, if it is similar to a
black body.
15 Planck's equation gives practically the same numerical values in the field of wave-
length and temperature with which we are here concerned.
3i8
TRANSACTIONS I. E. S. PART II
The black body's radiation having a peculiar importance in the
science of illumination, substances having an absorption law
expressed by the equation a = — |- B are peculiarly inter-
esting for the problem of heterochromatic photometry. Perhaps
it would be possible to find some other liquid fitting this law
better than does my blue solution, but that is a matter of chance.
Other means than absorption can be found to modify the intensi-
ties of different radiations (interference, chromatic or rotatory
0.024
0.022
0.020
0.018
0.016
t 0.0 14
o
<
o0.012
o
30.010
& 0.008
0.006
0.004
0.002
0
1.5 2 2.5^
Fig. 5.
polarization), and they have, as against absorption, this ad-
vantage that the law of alteration can be computed a priori; but
it does not seem possible to get by one of these methods the
law of opacity expressed by the above equation. The only com-
plete solution would be to separate the different simple radiations
in a spectrum, then weaken these differently by proper screens,
and finally recombine them. That is, of course, a very compli-
cated procedure, and the use of absorption, in spite of its theo-
retically imperfect character, seems to be the only way to solve
the problem practically.
\M
Ul
0
£
_z
V
1*
1
III
z:
■<.
0
Zj
_J
>-
Ul
Id
ft:
O
—J
CO
0
PROBLEM OF HETEROCHROMATIC PHOTOMETRY 319
DISCUSSION.
Dr. Herbert E. Ives (Communicated) : The Illuminating
Engineering Society has reason to feel honored by Prof. Fabry's
selection of the Transactions for the place of publication of
this admirable paper. His contribution is marked by a thorough
grasp of the subject, by much insight and ingenuity in choice of
the practical means for carrying out the central idea and as a
whole constitutes a very positive addition to the progress toward
practical colored light photometry.
The use of calibrated absorbing screens has, indeed, as Prof.
Fabry states, been suggested before, but such suggestions have
not been in the practical shape now proposed by him. They have
tacitly involved the calibration of every screen sent out from the
standardizing laboratory, not the calibration of a formula easily
reproducible by anyone. This constitutes a great advance. One
cannot but admire, too, the ingenious idea of making a universal
absorbing screen by applying the three-color principle in the
choice of absorptions. In general I should personally have been
suspicious of this method through fear that different observers
of different color vision would not find the standard screens to
perform their color difference eliminating function completely.
This objection is, however, very fully overcome by the choice for
the blue-green element of a medium which alone is sufficient to
perform by far the most usual transformation, namely, from
one black-body color to another, by producing very nearly an
exact spectral match, the same in appearance for all eyes.
Up to the publication of this paper it has happened that most
attention has been given, by workers in this field, to study aimed
at establishing methods for calibrating different colored standards.
As is clearly stated by Prof. Fabry, the actual practical use of
his absorbing screens is dependent upon the development and
adoption of such methods of calibration. In view of the recent
progress in clearing up the characteristics of different photo-
metric methods, Prof. Fabry's screens come before us at a most
opportune time.
It may be of interest to outline here the work along this line
which is now being done in the physical laboratory of the United
Gas Improvement Company, especially as the new absorbing
9
320 TRANSACTIONS I. E. S. — PART II
screens will take a prominent part therein. The work in general
consists in the calibration, by several different methods, of stand-
ards and absorbing screens for use in practical photometry. All
of these methods are developments of, and in conformity with,
the results of the writer's published work on heterochromatic
photometry. In short, the flicker photometer, under the condi-
tions specified in the writer's paper before the Society at the
1912 convention, will be the standard instrument, and the visual
luminosity curve determined by its use will be taken to represent
the normal eye. The first line of attack is through the use of a
special flicker photometer, now under construction, embodying
all the features called for by the previously mentioned work.
Auxiliary standards and absorbing screens will be calibrated
through observations by a large number of observers. The
problem of selecting absorbing media to be calibrated is obviously
enormously reduced by the timely appearance of Prof. Fabry's
work.
A second line of attack is by application of the spectrum
absorption curve of the various media to the normal visual
luminosity curve in the same manner that Prof. Fabry has indi-
cated. Great accuracy is aimed at. The spectral absorptions will
be determined by the aid of the photo-electric cell, which promises
to greatly excel the eye for sensitiveness in spectrophotometry.
In all cases of calibrating screens they will be used with the
standard 4-watt lamp as maintained at the Bureau of Standards,
perhaps also with the standard Pentane lamp and the Hefner.
A third method of attack is through the photo-electric cell of
one of the alkali metals, properly screened so as to have the sen-
sibility curve of the eye. Experiments toward this end are being
actively carried on.
It is only in one line of work that calibrated absorbing screens
might be found inconvenient, namely, in laboratory tests of incan-
descent mantles of varying compositions or electric incandescent
lamps of varying efficiency. To facilitate this kind of work it
is my present idea to use the special flicker photometer with
absorbing screens which will correct each observer's eye to nor-
mal. Study of a set of luminosity curves of 18 observers has
shown that practically all these observers can be corrected to
PROBLEM OF HETEROCHROMATIC PHOTOMETRY 32 1
normal by a yellowish or bluish screen placed over the eye.
Given two colors (such as solutions) which should measure
equally bright to a normal observer, and two solutions showing
absorption curves of long gradient, it should be possible for
every observer in a laboratory to be equipped with a screen to
carry on normal heterochromatic measurements.
All of this work will be reported in due course. It is men-
tioned here in order to indicate the prospects that before long
the measurement of different colored lights will be in practical
shape, both through the accurate calibration of Prof. Fabry's
ingenious screens and bv other means.
TRANSACTIONS
OF THE
Illuminating Engineering Society
Published monthly, except during July, August, and September, by the
ILLUMINATING ENGINEERING SOCIETY
General Offices: 29 West Thirty-Ninth Street. New York
Vol. VIII
OCTOBER. 1913
No. 7
Pittsburgh Convention.
The seventh annual convention of the
Illuminating Engineering Society was
held at the Hotel Schenley, Pittsburgh.
September 22 to 25, 1913. A program
of excellent papers, which occasioned
many spirited and interesting discus-
sions, a comparatively large attendance
and a generous complement of enter-
tainment combined to make this meeting
a pronounced success. In the history of
the society it marks another gain in the
rapid progress made by the society
within the seven years of its existence
in the promotion of the science and art
of illumination.
The registration totaled 459 members
and guests. It is probable, however, that
the attendance reached approximately
500. Of the latter number 176 were
members. The average attendance at
each session was approximately 100.
A prize, an ornamental desk lamp, was
won by the New York Section for hav-
ing the largest representation at the
meeting.
Several of the papers and reports
appear in this issue of the Transac-
tions; the rest will be published in the
November and December issues.
A brief outline of the convention pro-
ceedings is given in the following para-
graphs.
Monday, September 22.
The convention was called to order at
10 a. m. by the chairman of the Conven-
tion Committee, Mr. C. A. Littlefield.
An address of welcome was made by
Mr. William H. Stevenson, president of
the Pittsburgh Chamber of Commerce.
An appropriate response on behalf of
the society was made by Mr. Norman
Macbeth, vice-president.
A historic gavel and stand was pre-
sented to the society by Prof. George
Hoadley on behalf of the Philadelphia
Section. Inserted in the gavel, which is
of rosewood, are several contributions
or souvenirs, each one of which desig-
nates a stage or significant achievement
in the lighting industry. The stand con-
sists of a piece of a gas holder built in
Baltimore by the first gas company in
America, established in 1816. set in a
case of rosewood. A detailed descrip-
tion of the gavel and stand will appear
in another issue of the Transactions.
The address by President Preston S.
Millar surveyed the status of the present
day lighting conditions. Based on an
exhaustive canvass of numerous sources
of information, it constitutes a compre-
hensive appraisal with which future esti-
mates of progress in the science and art
of illumination may be compared. The
address will be published in a later issue
of the Transactions.
A report of the Committee on Organi-
zation of the International Committee
on Illumination was read by Mr. George
S. Barrows. The report reviewed the
TRANSACTIONS I. E. S.— PART I
movement started about two years ago
for the formation of an international
body to consider questions pertaining to
illumination. A meeting was held July 8
for the organization of a United States
National Committee and was attended
by delegates from the American Gas
Institute, American Institute of Elec-
trical Engineers, American Physical
Society and the Illuminating Engineer-
ing Society. Dr. E. P. Hyde, a past-
president of the I. E. S., was elected
president and Dr. C. H. Sharp secretary
of the United States National Commit-
tee. A meeting to reorganize the Inter-
national Photometric Commission was
held in Berlin, August 27 to 30, 1913.
Ten nations were represented by forty-
four delegates. The name of the Com-
mission was changed to the International
Illumination Commission and statutes
for its conduct were adopted. These
provide for the formation of national
committees by the national technical
societies interested in lighting. The
business of the Commission is to be
transacted by these committees, acting
through the honorary secretary of the
Commission, through delegates sent to
the general meetings which are to be
held every three years and through
standing committees. The next meeting
of the Commission is to be held in Paris
in 1916. A more detailed and formal
report on the proceedings and status of
the new Commission is to be received
shortly from the United States National
Committee and will be published in a
later issue of the Transactions.
The report of the Committee on
Progress was next presented by Mr.
F. N. Morton, chairman of the com-
mittee. The report appears in this issue.
At the afternoon session three papers
were presented : "Recent Improvements
in Incandescent Lamp Manufacture," by
E. J. Edwards and Ward Harrison;
"The Cooling Effect of Leading in
Wires upon the Filaments of Tungsten
Incandescent Lamps of the Street Series
Type," by T. H. Amrine; "Modern
Practise in Street Railway Illumination,"
by S. G. Hibben.
In the evening there was a reception
and dance in the ball-room of the Hotel
Schenley.
Tuesday, September 23.
Three papers were read during the
morning session : "The Psychic Values
of Light, Shade, Form and Color," by
F. Park Lewis, M. D.; "The Efficiency
of the Eye Under Different Systems of
Illumination, and the Effect of Varia-
tions in Distribution and Intensity of
Light," by C. E. Ferree of Bryn Mawr
College; and "Some Theoretical Con-
siderations of Light Production," by
W. A. Darrah. In connection with the
discussion of the latter paper, Mr. John
W. Howell of the General Electric Com-
pany gave a talk on current develop-
ments in the manufacture of incandescent
lamps. Several large high candle-power
nitrogen-filled, tungsten filament lamps
were exhibited. Mr. C. A. B. Halvor-
son, Jr., of the West Lynn Works of the
General Electric Company discussed the
status of the arc lamp, referring par-
ticularly to the magnetite lamp.
In the afternoon an inspection trip
was conducted to the works of the
Westinghouse Electric & Manufacturing
Company in East Pittsburgh.
The Section Development Committee
held a conference of the representatives
of section boards for the discussion of
section activities for the season begin-
ning October 1.
There was also a special meeting of
the Council in the afternoon. A report
of this meeting appears elsewhere in
this issue.
TRANSACTIONS I. E. S.— PART I
An evening session, to which the
public was invited, was held in the
Allegheny County Soldiers' Memorial
Hall. The lighting of this building,
which has been described in a paper in
the Transactions, was the subject of a
talk by Mr. Henry Hornbostel, the
architect of the building.
Mr. Georges Claude gave a lecture on
the neon tube lamp 01 which he is the
inventor, and showed several of the
lamps in operation. The lecture is in-
cluded in this issue of the Transac-
tions.
Two papers, "The Evolution of the
Lamp" by Roscoe E. Scott and "The
Quartz Mercury-vapor Lamp'' by W. A.
D. Evans were also presented.
Wednesday, September 24.
This day was designated as "Commer-
cial Day" and was devoted to a dis-
cussion of various lighting installations
involving the practical applications of
scientific lighting principles. The papers
discussed were : "Church Lighting" by
R. B. Ely; "Experiments in the Illumi-
nation of a Sunday School Room with
Gas" by E. F. Kingsbury; "Distinctive
Store Lighting" by C. L. Law and A. L.
Powell ; "Factory Lighting" by M. H.
Flexner and A. O. Dicker; "Store
Lighting" by J. E. Philbrick; "Hospital
Lighting" by W. S. Kilmer.
Wednesday afternoon the members of
the society were guests at a baseball
game between the Pittsburgh and
Chicago teams of the National League.
The annual banquet was held in the
evening. Prof. H. S. Hower of the
Reception Committee presided and intro-
duced the toastmaster General George
H. Harries. Addresses were made by
President P. S. Millar, President-elect
C. O. Bond, Charles M. Bregg and John
Brashear, "Pittsburgh's Grand Old
Man." Music was furnished by th>
Westinghouse Band.
Thursday, September 25.
At the morning session the followinj
papers were read : "The Lighting o
Show Windows" by H. B. Wheeler
"The Pentane Lamp as a Workins
Standard" by E. C. Crittenden and A. H
Taylor; "The Illuminating Engineering
Laboratory of the General Electrii
Company" by S. L. E. Rose; "Th<
Photo-Electric Cell in Photometry" fy
Prof. F. K. Richtmyer; "Some Studie:
in the Accuracy of Photometry" by E. J
Edwards and Ward Harrison.
At noon a number of the member;
attended a luncheon of the Jovians ii
the Hotel Schenley. Statesman F. M
Knapp of the latter organization pre
sided. Short talks were given by Pres
ton S. Millar, L. B. Marks, Normar
Macbeth, C. A. Littlefield and Georg<
Webster. Music was furnished by th<
orchestra of the Western Union Tele-
graph Company.
The afternoon session was opened bj
a paper entitled "Characteristics of En
closing Glassware" by V. R. Lansingh
The session was concluded by a genera
meeting of the society. The annua
report of the Council to the member
ship was presented by Joseph D. Israel
general secretary. The report will b<
published in a later issue of the Trans
actions.
The committee appointed at the firs
session of the convention to report 01
the presidential address then presentee
the following report :
Your Committee desires to commend the Re
port of your President for the careful perusal o
each individual member of the Society. As 1
record of the present status of the Art of Ilium
ination, it should assume an important place ii
the archives of the Society.
The classification of the elements of lightin;
practise into the so-called categories, is mos
useful not only in expressing present day con
TRANSACTIONS I. E. S. — PART I
ditions, but also as a means of evaluating future
progress. The careful and extensive survey of
lighting conditions through broadcast distribu-
tion of questions will undoubtedly yield much
important information.
Your Committee trusts that the President will
carry this investigation through to a conclusion
and make available such instructive data as may
be obtained.
A further important result may be anticipated
from this survey, that it will set to thinking a
large number of persons more or less interested
in the subject of Illumination, thereby stimulat-
ing activity on the subject.
To the Council it is desired to recommend for
consideration the President's suggestion that an
effort be made to secure commercial application
of the valuable information which is being con-
tinuously presented to the Society.
Your Committee believes it will be found de-
sirable to appoint a committee to investigate the
suggestion of the President, and recommend not
only with regard to carrying it out but also with
regard to the ways and means.
Respectfully submitted :
G. S. Barrows,
Norman Macbeth,
G. H. Stickney, (Chairman).
A Committee on Resolutions pre-
sented the following resolutions, which
were unanimously adopted :
Whereas, The Seventh Annual Convention of
the Illuminating Engineering Society is about
to terminate its sessions, and
Whereas, in general interest it has fully
maintained the high standard set by preceding
meetings of the Society, and
Whereas, this result has only been accom-
plished through the indefatigable effort and co-
operation of those charged with the responsi-
bility of the preparation for and the conduct of
the work. Be it Resolved :
That the thanks of the delegates be extended
To the Pittsburgh Section for their invitation to
visit Pittsburgh and their generous hospitality ;
To the General Convention Committee for the
comprehensive plan laid out by and so ably ac-
complished through their unceasing efforts ;
To the various Executive, Local and other
Committees who, by unselfish devotion to their
duties, have so largely added to the comfort and
entertainment of those in attendance ;
To the President of the Pittsburgh Chamber of
Commerce for his cordial welcome ;
To the County Commissioners for permission
to use Allegheny County Memorial Hall for a
session ;
To the Management of the Hotel Schenley for
the many courtesies extended to the delegates
and guests ; and
To the authors of the papers and the speakers,
without whose valued efforts the success of this
meeting would have been impossible ; and
Be it Further Resolved, that copies of these
resolutions be forwarded to the Pittsburgh Sec-
tion, to the President of the Pittsburgh Chamber
of Commerce, to the County Commissioners, to
the Chairmen of the various Committees, to the
Management of the Hotel, and that they be
spread upon the minutes of the Society.
President Millar then introduced
President-elect Charles O. Bond who
made a brief address in response to the
hearty reception which was accorded
him.
Mr. L. B. Marks introduced a resolu-
tion commending the work of President
Millar and his administration, which was
adopted unanimously. The convention
was then adjourned.
In the evening there was a theater
party at the Grand Opera House.
During the convention special enter-
tainment was provided for the ladies.
Monday afternoon there was a card
party at the Pittsburgh Athletic Club.
Tuesday a luncheon was served at the
factory of the H. J. Heinz Company;
there was also an automobile trip
through the city parks. Thursday after-
noon visits were made to the Carnegie
Institute and the Margaret Morrison
School.
In the way of sport for the men there
were several games of golf, and a tennis
game for a silver trophy which was won
by J. L. Wiltse.
Council Notes.
During the seventh annual convention
in Pittsburgh, a special meeting of the
Council was held September 23, 1913, in
the Hotel Schenley. Those present
were: Preston S. Millar, president;
George S. Barrows, C. O. Bond. Joseph
D. Israel, general secretary; V. R. Lan-
TRANSACTIONS I. E. S. — PART I
singh, Norman Macbeth, L. B. Marks,
treasurer; and \V. J. Serrill.
The Executive Committee reported
that it had transacted the following busi-
ness since the last regular meeting of
the Council in June. The report was
adopted.
Authorized the payment of vouchers Nos.
1359 to 1390 inclusive, amounting to $3600.48;
and vouchers Xos. 1391 to 1420 inclusive,
amounting to $1187.56.
An appropriation of $70 covering the cost
of programs, postage, etc., issued in conjunc-
tion with the joint meeting of the Illuminating
Engineering Society and the School Hygiene
Congress in Buffalo (August 25-3°. ,OI3), was
granted.
Elected twenty applicants to membership.
The names of these new members are given
on another page of this issue 01 the Trans-
actions.
Approved the appointment of the following
local representatives:
Abbott, A. L.
Manager, Northwestern Electrical Equip-
ment Company, St. Paul, Minn.
Collier, William Rawson.
Georgia Railway & Light Company,
Atlanta, Ga.
Hoar, F. Emerson.
Railroad Commission of the State of Cali-
fornia, 833 Market Street, San Francisco,
Cal.
Manahan, R. H.
City Electrician, Los Angeles, Cal.
Osborn, Fred. A.
University of \V ashington, Seattle, Wash.
Williamson, G. E.
Denver Gas & Electric Light Company,
Denver, Colo.
Approved appointment of two representa-
tives (Dr. E. P. Hyde and Mr. L. B. Marks)
from the Illuminating Engineering Society to
a meeting of representatives of several socie-
ties for the purpose of organizing a LTnited
States National Committee. The organization
of this committee is in line with a movement
started about two years ago for the formation
of an international body to consider questions
pertaining to illumination.
A draft of the annual report of the
general secretary was read. This report
with a few changes was made the report
of the Council to the membership of the
society.
An appropriation of $1,005 was
granted to cover expenditures to be
made by the Convention Committee in
connection with the Seventh Annual
Convention.
An appropriation of $35 was granted
to cover the cost of issuing notices
announcing a joint meeting of the
Illuminating Engineering Society and
the American Gas Institute during the
convention of the latter organization in
Richmond, Ya., in October.
The Finance Committee was directed
to engage a public accountant to audit
the accounts of the society for the
period January 1 to September 30, 1913,
at a cost not to exceed $50.
The following resolution was adopted.
Whereas, at this special meeting of
the Council of the Illuminating Engi-
neering Society, held in the Hotel
Schenley, Pittsburgh, Pa., September
23, 1913, during the period of the Sev-
enth Annual Convention of the Society,
it is deemed fitting and desirable that
the Council recognize the distinguished
services of the several committees which
have done so much to help the society's
work in support of the 1913 administra-
tion; therefore be it
Resolved, that the secretary be in-
structed to forward to the chairman of
each standing and temporary committee,
as well as to the chairmen of the con-
vention and sub-committees a copy of
this resolution, as an appreciation by the
Council of the committee work which
was so marked a feature of the activi-
ties of the society during the year 1913.
and as a recognition of the efficiency
with which the several committees per-
formed their various duties; be it
further
Resolved, that the Coun:il place on
record in this manner a renewal of its
TRANSACTIONS I. E. S. — PART I
expression of appreciation of the fidelity
and efficiency of the Assistant Secretary,
Mr. Joseph Langan, and his office assist-
ants.
It was voted to recommend to the next
Council the question of considering
ways and means of establishing the
responsibility of the society for all state-
ments made in papers presented by mem-
bers of the society at joint meetings with
other societies.
Prof. George D. Shepardson of the
University of Minnesota, Minneapolis,
Minn., was appointed a local represen-
tative of the society.
Mr. V. R. Lansingh, chairman of the
Committee on Sustaining Membership,
reported on the status of the sustaining
membership of the society.
It was voted to refer to the next
Council the question of supplying re-
prints and copies of papers to authors.
Vouchers Nos. 1421 to 1442 inclusive,
amounting to $691.20, were authorized
paid, subject to subsequent approval by
the Finance Committee.
The first regular Council meeting of
the administration 1913-1914 was held in
the general offices of the society, 29 West
39th Street, New York, October 10, 1913.
In attendance were: C. O. Bond, presi-
dent; Ward Harrison, J. D. Israel, gen-
eral secretary; V. R. Lansingh, C. A.
Littlefield, L. B. Marks, treasurer;
Preston S. Millar, W. J. Serrill, G. H.
Stickney.
The meeting was called to order by
President Bond at 10:40 a. m.
The minutes of a special meeting of
the Council, which was held September
23 in the Hotel Schenley, Pittsburgh,
Pa., during the Annual Convention,
were adopted.
Payment of vouchers 1443 to 1458 in-
clusive, amounting to $992.10, was
authorized.
Ten applicants were elected members.
Their names appear elsewhere in this
issue.
Eleven resignations from membership
were accepted.
Two applicants for sustaining mem-
bership were elected.
It was voted to hold the regular meet-
ings of the Council during the present
administration on the Friday following
the second Thursday of each month, at
10:30 a. m.
The following suggestions from the
preceding Council were referred to the
Finance Committee for recommenda-
tions :
(1) Set aside at the beginning of each year
for immediate investment in such securities as
may be recommended by the Finance Com-
mittee, a sum of money equal to 3 per cent,
of the preceding year's total revenue. To
provide for future contingencies of an extreme
character, some such provision seems war-
ranted.
(2) The moneys of the Society might be
deposited in a bank or trust company which
will allow interest on monthly or periodical
balances. The revenue from such a source
might easily exceed $100 annually.
The following suggestion from the
preceding Council was discussed and re-
ferred to the Committee on Progress for
recommendations :
With a view to making the Transactions
more valuable as a reference publication, a
list of current books and articles pertaining
to illuminating engineering might be published
each month in the Transactions. Were a brief
statement of the contents to accompany the
titles of these publications, this contribution
would be of so much greater value.
President Bond was directed to ap-
point a special committee to reconsider
the policy of the society in regard to
supplying authors with copies of their
papers which are published in the
Transactions. The present practise is
to give each author 50 of the advance
copies of his paper. When a paper is
not published in advance form, the
author does not receive any copies.
TRANSACTIONS I. E. S. — PART I
President Bond appointed the following
committee: G. H. Stickney, Chairman;
Dr. C. H. Sharp, Dr. Herbert E. Ives.
It was understood that the committee
would submit recommendations at the
November meeting of the Council.
The question of establishing the re-
sponsibility of the society for statements
made in papers presented by members
of the society at joint meetings with
other societies, and before meetings of
the Illuminating Engineering Society,
was referred to the Papers Committee
for recommendations, to be submitted at
the November meeting of the Council.
It was suggested that a note of some
sort might be placed upon each paper,
stating that the society does not neces-
sarily identify itself with the views or
opinions expressed in its papers.
An invitation from the American
Museum of Safety to the society, to join
in a session to be devoted to illumina-
tion, during an exhibition of the former
organization in New York, was accepted.
It was understood that the Papers Com-
mittee and the Committee on Reciprocal
Relations conjointly would designate
lecturers or authors to appear at the
session of the American Museum of
Safety, on the evening of December 18
in the Grand Central Palace, New York
City.
In connection with the latter meet-
ing, $25 was appropriated for litera-
ture to be issued in conjunction
with this session and a lighting exhibit
which is to be installed in the Museum
of Safety by a committee of the society.
It was understood that the exhibit would
also be displayed in the Grand Central
Palace during the Museum of Safety
Exhibition. This exhibit is to become
part of the permanent equipment of the
Museum of Safety.
Mr. G. H. Stickney, chairman of the
committee having charge of the installa-
tion of a lighting exhibit in the Ameri-
can Museum of Safety, reported briefly
on the work of his committee. He
stated that the expense of construction
and installation of the exhibit would be
borne by several lighting and manufac-
turing companies.
It was voted to donate to the School
Hygiene Congress a set of electros of
the illustrations in the illumination
primer "Light: Its Use and Misuse" and
a copy of the primer, with the under-
standing that it would be incorporated
verbatim and in its entirety in the pub-
lished proceedings of the Congress
which was held in Buffalo in August,
1913. The electrotypes are to be for-
warded to Dr. Thomas A. Storey of the
College of the City of New York, who
is secretary of the Congress.
Mr. Ward Harrison of the National
Electric Lamp Association of Cleveland,
was appointed a local representative of
the society in the city of Cleveland, Ohio.
The following local representatives
were reappointed for the present
administration :
Abbott, A. L.
Manager, Northwestern Electrical
Equipment Company, St. Paul, Minn.
Collier, William Rawson.
Georgia Railway & Light Company,
Atlanta, Ga.
Hoar, F. Emerson.
Railroad Commission of the State
of California, 833 Market Street,
San Francisco, Cal.
Manahan, R. H.
City Electrician, Los Angeles, Cal.
Osborx, Fred. A.
University of Washington, Seattle,
Wash.
Shepardson, G. D.
Professor of Electrical Engineering,
University of Minnesota, Minneapo-
lis, Minn.
8
TRANSACTIONS I. E. S. — PART I
Williamson, G. E.
Denver Gas & Electric Light Com-
pany, Denver, Colo.
The following committee chairmen
were appointed :
Progress :
F. E. Cady, National Electric Lamp
Association, Cleveland, Ohio.
Glare :
M. Luckiesh, National Electric
Lamp Association, Cleveland, Ohio.
Popular Lectures:
G. H. -Stickney, General Electric
Company, Harrison, N. J.
Sustaining Membership :
V. R. Lansingh, 6523 Euclid Avenue,
Cleveland, Ohio.
Advertising :
J. Robert Crouse, The National
Electric Lamp Association, Nela
Park, Cleveland, Ohio.
Lighting Legislation :
L. B. Marks, 103 Park Avenue.
New York City.
Section Development:
Joseph D. Israel, 1000 Chestnut
Street, Philadelphia, Pa.
Papers :
George H. Stickney, General Elec-
tric Company, Harrison, N. J.
It was resolved that the Council at
this time renew its expression of appre-
ciation of the excellent work done by
the Primer Committee in the prepara-
tion of the illumination primer "Light :
Its Use and Misuse" — the best piece of
constructive work the society has ever
accomplished.
The following Council Executive
Committee was appointed:
Charles O. Bond (ex-Ofiicio), Chair-
man,
3101 Passyunk Avenue, Philadelphia,
Pa.
Joseph D. Israel (ex-OMcio),
1000 Chestnut Street, Philadelphia,
Pa.
C. A. Littleeield,
55 Duane Street, New York City.
L. B. Marks (ex-OMcio),
103 Park Avenue, New York City.
Preston S. Millar,
80th Street and East End Avenue,
New York City.
Section Notes.
CHICAGO SECTION.
A meeting of the Board of Managers
was held in the Grand Pacific Hotel,
October 15. Those present were : Dr.
M. G. Lloyd, chairman ; J. R. Cravath,
vice-president; J. B. Jackson, secretary;
J. W. Pfeifer, C. C. Schiller, and H. B.
Wheeler.
The getting of suitable papers on car
lighting for the November meeting was
discussed. Dr. Lloyd and Mr. Jackson
were delegated to complete arrange-
ments for the meeting.
It was announced that Mr. W. A.
Durgin, assistant chief testing engineer
of the Commonwealth Edison Company,
would give a series of talks on the basic
principles of illumination before the sec-
tion meetings during the present season.
A proposed joint meeting with the
Railway Signal Association was dis-
cussed. The making of definite arrange-
ments for this meeting was withheld
pending further correspondence with the
association.
Committee appointments will be an-
nounced later.
A meeting of the Chicago Section was
held in the auditorium of the Western
Society of Engineers, Monadnock Block,
October 15. Vice-president J. R. Cra-
vath reviewed the proceedings of the
Pittsburgh convention. Two papers :
TRANSACTIONS I. E. S. — PART I
"Factory Lighting" by Messrs. M. H.
Flexner and A. O. Dicker and "The
Lighting of Show Windows" by H. B.
Wheeler were presented. Demonstra-
tions and exhibits of factory lighting
fixtures, show window reflectors and the
new nitrogen lamps were provided. A
report from the previous board on the
work of the past year was read by Sec-
retary J. B. Jackson. Dr. M. G. Lloyd
presented a tentative outline of the
meetings, papers, etc., for the present
year. Twenty-five members and five
guests were present.
NEW ENGLAND SECTION.
The Board of Managers of the New
England Section expect to announce
shortly a tentative program of papers
and meetings for the present season.
Mr. C. M. Cole, 156 Pearl Street,
Boston, Mass., has been appointed sec-
retary of the section.
NEW YORK SECTION.
The Board of Managers held two
meetings in the general offices of the
society, 29 West 39th Street, New York,
September 2 and 29 respectively. A
brief of the business transacted at these
meetings is given below :
It was decided that, whenever possible
during the present year, joint meetings
should be held with other societies.
Meetings with the following organiza-
tions were suggested : National Electric
Light Association, Sage Foundation for
the Prevention of Blindness, American
Society of Mechanical Engineers, New
York Electrical Society, National Com-
mercial Gas Association and the Munici-
pal Art Society. The secretary was in-
structed to prepare a budget for the
coming season. An effort will be made
to have issued some time in October or
early in November a printed tentative
program, listing papers, meetings, etc.,
for the entire season. Mr. A. S. Ives
was appointed a manager to succeed
C. R. Clifford, resigned. It was voted
to hold, as far as possible, an informal
dinner, immediately preceding each sec-
tion meeting, at Keene's Chop House,
70 West 36th Street, New York, on
the evenings of the section meetings.
Twelve applications for membership
were approved by the Board of Mana-
gers and transmitted to the secretary of
the society.
The following committee chairmen
have been appointed : Papers, H. V.
Allen; Membership, H. B. McLean;
Dinner, S. W. Van Rensselaer ; Attend-
ance, N. D. MacDonald ; Exhibition,
W. H. Spencer ; Reception, C. L. Law.
A meeting of the New York Section
was held in the United Engineering
Societies' Building, 29 West 39th Street,
New York, October 9. Mr. Norman
Macbeth reviewed the proceedings of
the seventh annual convention of the
society in Pittsburgh, in September.
A paper entitled "Distinctive Store
Lighting," which had been presented at
the convention, was presented by Messrs.
C. L. Law and A. L. Powell. The latter
paper was accompanied by a series of
lantern slides and autochrome color
plates demonstrating the advantages of
using color slides in photographic work.
Preceding the meeting there was an in-
formal dinner at Keene's Chop House,
70 Wrest 36th Street.
PHILADELPHIA SECTION.
The Board of Managers of the Phila-
delphia Section held two meetings, one
July i& and another September 5, 1913.
The following committee chairmen
have been appointed : Papers, H. A.
Hornor; Exhibition, H. Calvert; Mem-
bership, Samuel Snyder; Publicity, H.
10
TRANSACTIONS I. E. S. — PART I
H. Ganser; Dinner, F. C. Dickey;
Attendance, R. B. Ely.
A meeting of the Philadelphia Section
was held in the Engineers' Club, 1317
Spruce Street, Philadelphia, October 17.
A paper entitled "A Simple Unit Method
of Measuring Intrinsic Actinicity of
Flames and Surfaces for the Practise
of Photography" by Mr. F. M. Steadman
was presented by Prof. A. W. Good-
speed, of the University of Pennsylva-
nia. Vice-President W. J. Serrill re-
viewed the proceedings of the seventh
annual convention of the society in
Pittsburgh, September, 1913. During
the meeting there was an exhibition of
photographic lamps and lenses. About
60 members were present. An informal
dinner at the Engineers' Club preceded
the meeting.
The following program of meetings
and papers has been printed and issued
to the members of the section.
Friday, October 17.
Joint Meeting with Photographic
Society of Philadelphia.
"A Simple Unit Method for Measuring
the Intrinsic Actinicity of Flames
and Surfaces for the Practise of
Photography." By Mr. F. M. Stead-
man.
Presented by Prof. A. W. Goodspeed.
Address by Prof. Geo. A. Hoadley.
Report of Pittsburgh Convention.
By Mr. William J. Serrill.
Exhibition of Photographic Lenses and
Shutters.
Thursday, November 20.
Joint Meeting with Ophthalmologists of
Philadelphia at the College of
Physicians and Surgeons.
"Some Effects of Artificial Light upon
the Eye."
By Dr. Edward A. Shumway.
"The Effect of Mercury Arc Lamps
upon the Eyes."
By Dr. Geo. S. Crampton.
"The Lighting of a Private Library."
By Prof. Arthur J. Rowland.
Monday, December 8.
Joint Meeting with Philadelphia Section
A. I. E. E.
"Brightness Measurements versus Illu-
mination Measurements."
By Dr. Herbert E. Ives.
"Railway Car Lighting."
By Mr. Geo. H. Hulse.
"The Mercury Quartz Tube Lamp."
By Mr. Buckman.
Friday, January 6.
"Deficiencies of the Method of Flicker
for the Photometry of Lights of
Different Colors."
By Prof. C. E. Ferree.
Saturday, February 7.
Meeting under the Auspices of Drexel
Institute.
"Light and How to Use It."
By Mr. C. O. Bond, President
of I. E. S.
Wednesday, February 18.
Joint Meeting with Franklin Institute.
"Artificial Daylight."
By Dr. Herbert E. Ives.
Friday, March 20.
"Lighting and Signalling Systems of
Subways."
By Mr. F. D. Bartlett.
"The Sun — The Master Lamp."
By Prof. James Barnes.
Thursday, April 9.
Joint Meeting with Franklin Institute.
"Recent Developments in the Art of
Illumination."
By Mr. Preston S. Millar.
TRANSACTIONS I. E. S. — PART I
11
Friday, April 17.
"The Structure of the Normal Eye and
its Ability to Protect Itself Against
Ordinary Light."
By Dr. Wendell Reber.
"Glassware for Illumination and Other
Purposes."
By Mr. James Gillinder.
Friday, May 15.
Mass Meeting of all the Engineering
Societies of Philadelphia and
Vicinity.
Special Program to be arranged and to
include an address on
"The Relation of Engineers to the
Progress of Civilization."
By Dr. Chas. Proteus Steinmetz.
PITTSBURGH SECTION.
The Board of Managers of the Pitts-
bugh Section has announced the follow-
ing tentative program of meetings and
papers for the present season :
October — "Technical Elements of Vi-
sion" by Dr. H. H. Turner.
November — -"Technical Discussion of
the Elements of Lighting" by Prof.
Hower and others.
December — A joint meeting with the
Pittsburgh Section of the American
Institute of Electrical Engineers. A
Central Station paper will be pre-
sented by H. N. Muller of the
Duquesne Light Company.
January — A paper to be selected by the
members from Cleveland. The sub-
ject will be announced later.
February — "Railroad Yard Lighting" by
A. C. Cotton and A Kirschberg of
the Pennsylvania Railroad Com-
pany.
March — A "gas lighting subject; the
speaker to be announced later.
April — "Developments of Flame Carbon
Arc Lamps" by C. E. Stephens.
May — "Store Lighting" ; speaker to be
announced later.
June — Open.
The following committee chairmen
have been appointed : Papers, C. E.
Stephens ; Exhibition, S. G. Hibben ;
Membership, C. H. Mohr; Publicity,
M. C. Turpin; Dinner, R. H. Skinner;
Reception, V. R. Lansingh; Education,
W. A. Darrah.
New Members.
The following applicants were elected
members of the society September
1913:
Airston, Alexander.
Head of Factory Lighting Depart-
ment, Westinghouse Electric & Mfg.
Company, East Pittsburgh, Pa.
Blumenauer, C. H.
President and Treasurer, Jefferson
Glass Company, Follansbee, W. Va.
Bowen, Dudley A.
Salesman, Westinghouse Electric &
Mfg. Company, 165 Broadway, New
York, N. Y.
Brand, Walter C.
Illuminating Engineer, Macbeth-
Evans Glass Company, Wabash
Building, Pittsburgh, Pa.
Bullard, John E.
Manager New Business Department,
Public Service Corporation of L. I.,
Port Washington, N. Y.
Cadby, J. N.
Chief Inspector of Electric Service,
R. R. Commission of Wisconsin,
Madison, Wis.
Damron, C. E.
Assistant in Commercial Engineer-
ing Department, General Electric
Company, Harrison, N. J.
Hake, Harry G.
Assistant Professor of Electrical
Engineering, Washington Univer-
sity, St. Louis, Mo.
12
TRANSACTIONS I. E. S. — PART I
Harrison, Haydn T.
Electrical and Illuminating Engi-
neer, ii Victoria Street, London,
S. W., England.
Langworthy, E. L.
Eastern Manager, The Adams &
Westlake Company, 2218 Ontario
Street, Philadelphia, Pa.
McNary, S. J.
Vice-President and Illuminating En-
gineer, The E. G. Jones Electric
Company, 141 East 4th Street, Cin-
cinnati, Ohio. lyS
Pascoe, C. C.
Manager of Lamp Sales and Illumi-
nation Department, General Electric
Company, 185 Paseo Colon, Buenos
Aires, Argentine Republic.
RilEy, F. M. H.
Lighting Specialist, Kansas City
Electric Light Company, Grand
Avenue, Kansas City, Mo.
Stannard, Clare N.
Secretary, Denver Gas & Electric
Light Company, Denver, Colo.
Swallow, Joseph G.
Chief of Installation and Inspection,
United Electric Light & Power
Company, 11 70 Broadway, New
York, N. Y.
Trotter, A. P.
Electrical Engineer and Advisor,
Board of Trade, Whitehall, London,
S. W., England.
Voyer, Leonard E.
Assistant in Illuminating Engineer-
ing Laboratory, General Electric
Company, Harrison, N. J.
Walters, G. L.
Sales Manager, The Adams & West-
lake Company, 319 West Ontario
Street, Chicago, 111.
Watkins, Frederick A.
District Sales Manager, Pittsburgh
Reflector & Illuminating Company,
565 West Washington Street, Chi-
cago, 111.
Williamson, G. E.
Denver Gas & Electric Light Com-
pany, Denver, Colo.
The following applicants were elected
members October 10, 1913 :
AtmorE, A. L.
Engineering Department, Philadel-
phia Electric Company, 1000 Chest-
nut Street, Philadelphia, Pa.
Barnes, James.
Professor of Physics, Bryn Mawr
College, Bryn Mawr, Pa.
Federer, Theo. P.
Incandescent Lamp Salesman, Gen-
eral Electric Company, 30 Church
Street, New York, N. Y.
Fortune, F. R.
Lighting Specialties Company, 650
Century Building, Pittsburgh, Pa.
Gray, Henry J. W.
Philadelphia Electric Company, 1000
Chestnut Street, Philadelphia, Pa.
Loeouist, Harry S.
Engineering Department, National
Lamp Works of General Electric
Company, Nela Park, Cleveland, O.
McQuiston. J. C.
Advertising Manager, Westinghouse
Electric & Mfg. Company, East
Pittsburgh, Pa.
Moon, T. Elmer.
Illuminating Engineer, James ' E.
Hamilton Engineering Co., 1625
Real Estate Trust Building, Phila-
delphia, Pa.
Strait, E. N.
Public Utility Expert, Railroad
Commission of Wisconsin, Madison,
Wis.
Turpin, M. C.
Westinghouse Electric & Mfg. Com-
pany, East Pittsburgh, Pa.
/
TRANSACTIONS I. E. S. — PART I
13
Sustaining Members.
The following additional applicants
have been elected sustaining members of
the society :
Alexalite Company.
432 East 23rd Street, New York.
N.Y.
Board of Water Commissioners.
City Hall, London. Ontario, Canada.
Cooper Hewitt Electric Company.
Eighth and Grand Streets, Hoboken,
N.J.
Edison Electric Illuminating Co. of
Brockton.
42 Main Street, Brockton, Mass.
H. W. Johns-Man ville Company.
Madison Avenue and 41st Street,
New York.
Little Rock Railway & Electric Com-
pany.
IIS West 4th Street, Little Rock,
Ark.
Pittsburgh Lamp, Brass & Glass Co.
Pittsburgh, Pa.
The Leeds & Northrup Company.
4901 Stenton Avenue. Philadelphia,
Pa.
Local Representatives.
To extend the influence and work of
the society into cities and territories in
which there are not sections of the
society the Council has recently ap-
pointed eight local representatives. The
names of the representatives are listed
on a page in the front part of this issue.
These representatives will communicate
to the general secretary from time to
time information concerning local devel-
opments in which the society is con-
cerned. They will also endeavor to pro-
mote occasional meetings under the
joint auspices of the I. E. S. and local
organizations.
TRANSACTIONS
OF THE
Illuminating
Engineering Society
OCTOBER, 1913
PART II
^
*r
Papers, Discussions and Reports
[ OCTOBER, 1913 ]
CONTENTS -- PART II
Report of the Committee on Progress 323
Psychic Values of Light, Shade, Form and Color. By
F. Park Lewis, M. D 357
Neon Tube Lighting. By Georges Claude 37 *
The Illuminating Engineering Laboratory of the General
Electric Company. By S. L. E. Rose 379
y>
REPORT OF THE COMMITTEE ON PROGRESS.*
To the Illuminating Engineering Society:
During the past year the science of illumination has probably
made greater progress than at any other similar period of its
history. While few radical changes or developments have been
made in connection with light sources, improvements have been
made in mechanical construction of present systems, resulting in
increased efficiency, and illumination has become the subject of
study by physicists, oculists, architects, legislative bodies and
others as never before.
INCANDESCENT ELECTRIC LAMPS.
In connection with light sources the development of the
tungsten lamp stands out most prominently. This, in high candle-
powers, is a serious competitor of the arc lamp both for indoor
and outdoor lighting, and has largely superceded the carbon fila-
ment lamp, as shown by the following table taken from a recently
published article, and giving the relative number of lamps sold.1
Type
Per cent.
Per cent.
Per cent.
Per cent.
Per cent.
Per cent.
1907
1908
1909
1910
191 1
1912
Carbon . . .
93-27
84.12
68.98
63.08
52.90
25.47
Gem
5-88
8.58
I5-07
14.88
19.00
33-59
Tantalum
o-75
I.78
2.12
3-57
2.74
1. 00
Tungsten.
0. 10
5-52
I3.83
IOO.OO
18.47
25-30
39-94
Total ....
T OO.OO
100.00
100.00
99-94
100.00
It is extremely probable that more recent figures will show that
the tungsten lamp, aided by greatly reduced price, its high effi-
ciency and the fact that electric companies are beginning to make
free renewals of tungsten lamps as they did of the carbon fila-
ment, is destined to make the carbon filament lamp a vanishing
quantity. In this connection it is interesting to note that the
United States government has issued an order through the office
of the supervising architect of the Treasury Department that
carbon and metallized carbon-filament lamps are not to be used
* A report read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
1 Lighting Journal, July, 1913.
324 TRANSACTIONS I. E. S. — PART II
in any of the government buildings, and any such lamps in use
at the time of the receipt of the order must be removed and 25-
watt tungsten lamps substituted.2
There has been marked improvement in the life of some of the
higher wattage 100-volt types and in compensator and train light-
ing lamps. In the 60, 100 and 150-watt, 100-volt lamps improve-
ments in life have taken place similar to those made last year in
the cases of the 250, 400 and 500-watt lamps. This improvement
in quality is on the order of 50 per cent, in life and has made
possible substantial improvements in efficiency. The use of chem-
icals— the so-called "vacuum getter," in the bulbs of these lamps
for reducing the blackening of globes has largely brought about
these improvements.3' 4 At present the commercial efficiencies of
these types are around 1.17 w. p. c. while the higher wattage types
are put out at about 1 w. p. c.3> 4
To meet a demand for a 60-watt small base lamp for use in
residence lighting and to provide a 40-watt lamp that will in all
cases be interchangeable with the 25-watt lamps these two types
are now made in smaller bulbs. The 60-watt lamp is now made
in the same bulb as was standard for 40-watt lamps and the 40-
watt lamp in the bulb which is now standard for 25 watts. This
change has been made without impairing the life or efficiency of
the lamps.3
Improvements in the tungsten wire-drawing process have made
it possible to manufacture wire of almost exactly the desired size
and so render it possible for manufacturers to make lamps of so
nearly the desired voltage and efficiency that the function of
photometry in the lamp factory is now principally to guard
against errors in manufacture. This improvement has been
adopted by most of the manufacturers in this country for all
lamps intended for series burning service and has been productive
of greatly improved results in all cases.3
A 10- watt, 100 to 130-volt lamp and a 5-watt, 50 to 60-volt
lamp have been developed. These are important additions to
the line of sign lamps in that they furnish a low wattage lamp
2 Electrical World, March 15, 1913.
3 Electiical Journal, June, 1913.
4 Electrical Journal, January, 1913.
REPORT OF THE COMMITTEE ON PROGRESS 325
that does not require the operation of more than two lamps in
series upon 100 or 200-volt circuits.
Miniature lamps of the candelabra and decorative types with
tungsten filaments have been developed for no-volt circuits.
These have the filaments wound to form a helical coil of small
diameter and this helix is then mounted on the supports in the
same manner as an ordinary filament. The mandrel around
which the filament is formed is so small that the helical con-
struction of the filament can scarcely be recognized with the naked
eye. This style of construction renders possible the use of fewer
supports and smaller bulbs than ordinary tungsten lamp construc-
tion would allow.3
During the past year the use of concentrated filament tungsten
lamps has enormously increased. At the present time a very
large proportion of the automobiles now being sold in this coun-
try are equipped with these lamps. Concentrated filament lamps
are also being used for stereopticon work, for trolley car head-
lights and for theatre stage lighting in place of arc lamps and
Nernst lamps. With the style of construction now used excellent
concentration is obtained; the filament for a ioo-watt, 115-volt
lamp occupies a space only 7/16 inch (11.113 mm.) in diameter
and 7/16 inch long.
A new filament consisting of an alloy of tungsten has also been
brought out. It is claimed that this not only possesses the strength
and efficiency of drawn tungsten, but that it will withstand crys-
tallization for a longer period.0
An investigation has been made upon the heating of screw
thread lamp sockets. This was found to be almost entirely due
to the heat dissipated by the lamp or radiator and has little to do
with the energy lost in contacts. Bayonet holders are satisfactory
up to 250 watts, while screw sockets can transmit energy up to
1,000 watts with satisfaction.6
Possibly the most startling development in the tungsten lamp
is the announcement of one giving an efficiency of 0.5 watt per
candle. A specially shaped tungsten filament is used in a bulb
containing an inert gas, as nitrogen, at a pressure of about one
6 Electrical World, December 7, 1912.
6 Electrician, June 27, 1913.
326 TRANSACTIONS I. E. S. — PART II
atmosphere. The types to be developed first are adapted to high
current consumption, say 6 amperes and over.7
A tungsten lamp has recently been brought out which gives a
color very closely resembling that of daylight. It is not, how-
ever, an exact equivalent, and is, therefore, not suited to color
matching. The energy consumption of this lamp is about 1.4
watts per candle, and it has a candle-power of from 70 to 75.8> 9
A method of testing for spots in incandescent lamp filaments
lias been developed in which the body of a filament is made to
disappear against a luminous background until the light or dark
spot shows by contrast.10
Experiments were made to determine the variation in candle-
power produced by frosting. The distribution curves were found
to be very much the same before and after frosting, and half-
frosting resulted in a reduction of from 10 to 20 per cent, in
the candle-power. One fact brought out in these experiments was
of practical use in permitting a close estimate to be made of the
efficiency of a half frosted bulb. It was found that the curves
for the distribution of light of each lamp, clear and half frosted,
intersected at an angle very close to 66° in each case, and the
candle-power at this angle is in each case very nearly 45 per cent.
of the horizontal candle-power of the clear lamp. Thus the
horizontal candle-power of the clear bulb may be obtained by
multiplying that obtained with the half frosted bulb at 66° by
2.22.11
A focusing tungsten lamp with a spherical bulb silvered near
the socket has been introduced within the past year. This lamp,
which is rated at 32 candle-power has an efficiency of 1 watt per
candle. In another lamp there is a specially shaped reflector
in close proximity to the filament. The leading-in wire enters
at the apex of the reflector and leaves at its circumference.12
A system has been developed in which 14-volt lamps are used
with a special transformer for each lamp or for each group
of lamps. Each transformer is connected in parallel to the line,
' Electrical World, July 19, 1913.
8 Zeit.furBcleucht., April 30, 1913.
9 Elek. Anzeig., April 24, 1913.
10 Electrical World, May 3, 1913,
11 Electrician, April 25, 1913.
1- Elek. Anzeig., April 24, 1913.
REPORT OF THE COMMITTEE ON PROGRESS 327
and when it is disconnected the primary circuit is broken so
that there are no no-load losses. These lamps are cheaper than
the standard no-volt lamps, in addition to which advantage are
the long life and the possibility of using low candle-power.
The lamps are thought to be especially suitable for agricultural
districts.13
In order to avoid the excessive glare inherent in filament
lamps, a patent has been taken out according to which the outer
surface of the lamp bulb is made in a series of fine substantially
parallel grooves constituting a system of prisms. The grooves
may be arranged in various ways.14
ELECTRIC ARC LAMPS.
The open and enclosed series and multiple arc lamps are fast
disappearing, their places being taken by the magnetite and the
flame arc lamps. There seems to be during the past year, as far
as arc lamps are concerned, a strong tendency towards the use
of larger units and the cutting down of the intrinsic brilliancy
of the arc even at a sacrifice of efficiency.
The flame arc lamp is of special value in lighting large areas,
and is particularly adapted to smoky and dusty places, such as
foundries, blacksmith shops, and railroad train sheds. In fact
one of the largest railroads in the country spent months in trying
out various systems of lighting for the train shed of one of their
large stations, and, after experimenting with various systems of
lighting finally decided to use the flame arc lamp exclusively,
adopting it also for the yard.
This lamp has been improved so that the fumes thrown off
by the arc are condensed and so are prevented from forming a
deposit on the globe and from escaping into the air. The lamp
is also economical as far as maintenance is concerned, as one
set of carbons will burn from 100 to 120 hours without attention.
A rather novel device is an arrangement for converting any
enclosed arc lamp, alternating or direct current, series or mul-
tiple, and regardless of voltage, into a flame arc lamp.15 Many
minor improvements have been made in construction of the arc
18 Electrical World, December 2S, 1912.
14 Electrical World, September 14, 1912.
lb Electrical World, March 1, 1913.
328 TRANSACTIONS I. E. S. — PART II
lamp such as making the clutch work on the electrode indirectly,
thereby keeping it independent of the size of the electrode, im-
proving the feeding arrangement, etc. The lamps are now made
to give any desired range of color and to work under any ordi-
nary condition even, as in one lamp recently introduced, on a
25 cycle circuit, and have reached an efficiency of 0.25 watt or
better per candle with possibilities of further improvement.
One of the most interesting applications of the long-hour
series alternating current flame carbon arc lamp is to low fre-
quency circuits of 25 cycles. With the unimpregnated carbon
lamp, there is a flicker at each reversal of polarity which, at
this frequency, is very marked. In the flame arc lamp, the light
is obtained from the arc screen and is, therefore, independent
of polarity.16
A discussion was reported about the end of last year in which
were reviewed the results produced upon the stability of the
electric arc by wind currents, magnetic fields, movements and
presence of solid obstruction, etc.17 In an investigation made
in regard to the evaporation temperature of the arc lamp it was
found that the temperature is independent of the current, but
varies with atmospheric pressure, proving that the atmosphere
of the crater is at the boiling temperature of carbon. A peculiar
phenomenon noticed was that the positive crater seems to begin
to boil at pressures below atmospheric.18
VACUUM TUBE LAMPS.
One of the more recent developments of the mercury vapor
lamp is a combination with the tungsten, an apparent white light
being thereby produced. The unit is very compact and is fur-
nished with a novel starting device. The consumption of this
lamp is claimed to be 0.73 watt per mean hemispherical candle-
power.19 Another recent form of the mercury vapor lamp is one
with a specially designed quartz tube and designed particularly
for street lighting. This lamp is started by means of a heated
spiral which vaporizes a small portion of the mercury; this is
automatically cut out of circuit when the lamp is in operation,
16 Electrical World, July 26, 1913.
17 Electrical Journal, December, 1912.
18 Jnl.fur Gasbel., July 12, 1913.
19 Illuminating Engineer (L,ondon), October, 1912.
REPORT OF THE COMMITTEE ON PROGRESS 329
and renders tipping unnecessary. The lamp is recommended for
the varying voltages met with on traction circuits and is said
to consume about 0.4 watt per candle.20- 21
Another development has been the use of tungsten instead of
platinum wires for sealing in of the electrode wires and the em-
ployment as a sealing material of a special high temperature
glass, having nearly the same coefficient of expansion as the
tungsten. Graded glass is inserted between the sealing glass
and the quartz chamber walls in order to effect a suitably graded
transition from the quartz to the sealing material.22
A new method has been devised, applicable to any lamp de-
pending upon a vacuum, of sealing the conductors through the
glass, and by which perfect seals may be obtained between easily
oxidizable metals and glass of low fusing point. After inserting
the conductor through the aperture in the glass, the latter is
strongly heated by means of a blow pipe flame until perfect co-
hesion has been obtained between the glass and the metal. The
seal is then taken out of the flame, and when it reaches a dull
red, the leading-in wire and the glass surrounding it are cooled
by several immersions in a special bath, say of sperm or other
oil, wax or fat.23
The value of the ultra-violet rays of the quartz tube mercury
vapor lamp is becoming recognized. The lamp is now used for
sterilizing and the destruction of bacteria. This effect, however,
decreases in time, due partly to the formation of an obscuring
deposit.24
Announcement has also been made of a vapor lamp giving a
white light. In this the tube is filled with the vapor of cadmium
with from 3 to to per cent, of mercury. At its most economical
point the specific consumption is 0.16 watt per candle.25
A new source of light for photo-electric work and said to
have an effect 250 times as great as the mercury lamp has been
introduced. This is a vacuum discharge tube containing hydrogen
at low pressure.26
20 Electt otech. Zetlschr., March 20, 1913.
81 Elek. Am., February 13, 1913.
■ Electrical World, May 10, 1913.
23 Electrician, July 4, 1913.
-* Soc. Int. Elect, Bull. 2, Ser. 3, June, 1912.
26 Electrotech. Zeitschr., Septembers, 1912.
20 Physical Review, April, 1913.
33° TRANSACTIONS I. E. S. — PART II
Various investigations have been made upon the characteris-
tics of neon lamps. It was found that when the tube was placed
in an alternating current 4,000-volt circuit with a transformer
and condenser, 0.72 amperes usefully traversed the tube Intro-
duction into the circuit of more neon tubes, instead of decreasing
the current, apparently increased it up to four tubes, after which
the current diminished. The phenomenon was not explained,
but it was evidently not due to resonance.27
Disappearance of various gases by passing a discharge for
some time through vacuum tubes was the subject of another
investigation and was found to be due to definite chemical action
rather than to physical absorption.28
Investigation made also of the cause of the short life of neon
tube lamps showed that it was due to absorption of neon by
the electrodes, and that if the current density at the electrodes
was small, the life of the lamps would be greatly increased. By
employing very large electrodes, therefore, tubes showed no
deterioration in 2,000 hours' use.29' 30
GAS LAMPS AND APPURTENANCES.
A large number of new types of gas lighting units have made
their appearance in both hemispheres. Some of them have been
designed with a view to more universal adaptability, such as a
burner with a curved bunsen which may be turned either up or
down and is therefore, applicable to either upright or inverted
burners ;31 some with a view to ease of maintenance, such as one
in which the main feature is the possibility of lifting from the
fixture, without the use of pliers ;32 and some with a view to
increased efficiency, such as one in which the mixture of gas
and air issue at high velocity;33 resulting in an efficiency claimed
to be double that of the standard type; also another lamp with
an upright bunsen and an inverted mantle.34 Low-pressure high-
efficiency lamps, particularly for out-door use and giving 1,500
27 Le Genie Civil. December 14, 1912.
23 Electrician, November 15, 1912.
29 Electrician, May 30, 1913.
30 Comples rendus, April 28, 1913.
31 Journal of Gas Lighting, October 1, 1912.
32 Journal of Gas Lighting, September 24, 1912.
33 Le Genie Civil, November 23, 1912.
34 Journal of Gas Lighting, September 14, 1912.
REPORT OF THE COMMITTEE ON PROGRESS 33 1
candle-power and with an efficiency of nearly 40 candles per
foot have been introduced abroad, but have not yet appeared in
this country.
What is probably the most important development in gas
lighting during the past year is the introduction of a high-power
single-mantle inverted lamp filling a place between the old small
unit and the multiple mantle lamp. The lamp gives from 180 to
225 mean spherical candles according to the gas used and the
pressure, and thus occupies a space hitherto unfilled. A valuable
feature of this lamp is the fact that, because of the single mantle
the light is concentrated and reflectors and glassware may be
designed with much greater precision than with the multiple
mantle lamp.35
Hitherto one of the difficulties in connection with the use of
high-pressure lamps was the heat developed which limited the
kinds of glassware available and also necessitated a globe of very
great size. In a modification of one of the high-pressure inverted
lamps now in wide use, particularly in England, a small silica
cup about 5 inches high, and therefore but little larger than the
mantle itself, has been substituted for the globe hitherto used.
This cup totally excludes the secondary air supply, all air for
combustion being admitted as primary air. The lamp, moreover,
is greatly reduced in size and the reflector is omitted. It is
claimed that after allowing for the absorption of light by the
silica cup a 10 per cent, increase in efficiency is obtained. These
cups are made for lamps of from 60 to 1,500 candle-power.
They are very durable and will not break if the mantle gives
way and the flame strikes them.36- 37
A new high-pressure lamp has also been developed in the
United States, designed particularly for street lighting, and
working under a pressure of 55 inches of water or 2 pounds
per square inch. These lamps are at present made in two sizes —
500 and 1,100 candle-power respectively and have an efficiency
of over 50 candles per foot of gas per hour. The mantles are of
artificial silk and have a life of over 400 hours with practically
no depreciation in candle-power.38 High pressure lighting, how-
85 Lighting Journal July, 1913.
38 Journal of Gas Lighting, February 18, 1913.
3: Journal of Gas Lighting, February 25, 1913.
" American Gas Light Journal, December 30, 1912.
332 TRANSACTIONS I. E. S. — PART II
ever, has not been developed in the United States to the same
extent as abroad.
The artificial silk mantle is steadily coming into use, its dur-
ability, long life and strength being its recommendations. By a
recent improvement in this type of mantle, it can be burned di-
rectly on the burner, so that all burning off, collodionizing, etc.,
becomes unnecessary.39
A most interesting and valuable investigation was made to
determine the cause of the falling off in candle-power of the
incandescent mantle. Microscopical examination showed that
when new the ash seemed to be made up of very small particles,
making an opaque and light giving structure. As the mantle was
heated, these particles gradually fused together to form larger
particles and, by degrees a large transparent mass, which, by
the laws of physics, is comparatively without light giving prop-
erties.40
Other experiments were made to apportion to each part of
the lamp its share of the deterioration. Under favorable con-
ditions, the deterioration in 1,000 hours' continuous burning was:
Per cent.
For the mantle alone zyi
For the burner alone 2>£
For the glassware alone 10
Total deterioration 15
of which, by proper maintenance, 12^ per cent, could have been
eliminated.41
Automatic gas ignition is having a steady growth. Pilot lights
are increasing in number, while the jump-spark and filament
igniters are gradually coming into use. Distance control by
means of electrically operated gas cocks and also pneumatic con-
trol in connection with pilot and jump-spark ignition are being
extensively employed, particularly in the past year. It should
be noted here that pilot lights have been improved by adopting
the bunsen principle and by better protection, so that the danger
of extinction is greatly reduced. Self-kindling mantles are also
receiving attention, and one improvement recently made in this
type of mantle was to mix ammonium chloride with rhodium
33 Gas World, February 1, 1913.
40 Lighting Journal, April, 1913.
« Transactions Illuminating Engineering Society, December, 1912.
REPORT OF THE COMMITTEE ON PROGRESS 333
chloride. The metallic rhodium deposited is of a silver grey
and comes to full intensity without presenting a dark streak down
the mantle. It has since been found that lithium chloride is
cheaper and works better.42
The pyrophoric alloys are also in process of development, and
are now on the market. A new composition of the alloy has been
found which, instead of giving a shower of sparks, produces a
long flame, thus rendering the friction wheel unnecessary. This
alloy, is air proof, and does not disintegrate like the old cerium
alloy.43
ACETYLENE.
The use of acetylene is increasing for isolated buildings and
for distribution from a central plant in very small towns. A
large number of burners are available to suit varying needs, but
as yet no incandescent burner has been devised for use with
acetylene that is thoroughly satisfactory. An enormous field for
acetylene has been developed in lights for automobiles, where
acetylene dissolved in acetone is dividing honors with the
tungsten lamp. It is also stated that acetylene mixed with oxy-
gen and used on a special disk of ceria gives a light almost
equal to the electric arc and will shortly be put to use with the
cinematograph.44
SUNDRY SYSTEMS.
Air gas systems and systems employing kerosene under pres-
sure for use in connection with incandescent mantles are mak-
ing considerable progress in the rural districts and other places
not reached by city gas and electric plants. The former system,
however, has had a setback in the increased price of gasoline and
the difficulty in obtaining the high grade necessary. A new
modification of the latter system consists of a gas arc lamp con-
nected with the fuel tank by a fine flexible bronze tube. Labora-
tory tests showing 1,220 candle-power are claimed for this
lamp.43
A new method of utilizing natural gas has recently been
*- Gas World, February 15, 1913.
43 Journal of Gas Lighting, September 17, 1912.
44 Gas IVorld, January 18, 1913.
4f> /run Age, January 23, 1913.
334 TRANSACTIONS I. E. S. — PART II
brought out, but is still in an experimental stage. The plan is to
pump the nearly exhausted wells and treat the gas in such a
way as to produce a volatile liquid which, on release of pressure
would evaporate into a utilizable gas.
COLD LIGHT.
The startling announcement has been made that the long
looked for secret of obtaining a cold light has been discovered in
France, it being understood that the term "cold" is purely rela-
tive. A rotating wheel carrying a number of lamps at the cir-
cumference is arranged so that an electrical contact lights each
lamp in succession. The result is a series of instantaneous
flashes as each lamp passes the contact point, retinal persistance
giving the effect of a steady light. With this apparatus, the
inventor claims to be able to run the lamps at a voltage of from
two to four times the normal, and to obtain an efficiency thereby
of 0.2 watt per candle, 2 watts per candle being normal.46 In
this connection, it may be said that the idea of supplying lamps
with intermittent current is not new. Some fifteen years ago an
inventor placed in the lamp circuit a commutator by which the
current was interrupted a number of times a second, but nothing
ever came of the plan. A test of the present system is on record
from which we learn that the results are what might be ex-
pected from overrunning the voltage; the life of the lamps was
reduced from one thousand hours to four hours.47
LUMINOUS ANIMALS.
The firefly is still the subject of research among scientists.48
It is pretty well known that certain animals and insects have the
power to produce by means of chemical reactions a form of
radiant energy lying practically within the limits of the visible
spectrum, but the nature of these reactions is not yet known.
One investigator, indeed, separated out two substances called
by him eluciferase and luciferine respectively which, when mixed
and moistened, became luminous.49 The nature of these sub-
stances is, however, a matter of doubt. A study has been made
40 Scientific American, May 31, 1913.
*i Journal of Gas Lighting, May 13, 1913.
48 Lighting Journal, January, 1913.
« Philadelphia Pubic Ledger, April 6, 1913.
REPORT OF THE COMMITTEE ON PROGRESS 335
of the intrinsic brilliancy of the glowworm or larva of the fire-
fly, which gave as a result 0.0046 candle per square centimeter.
From this it was calculated that the luminous material of the
glowworm, could it be reproduced, would not only be of high
luminous efficiency but would also be a happy mean in intrinsic
brightness, far lower than the artificial light sources with which
we now try our eyes, yet high enough to permit its use without
pre-empting more wall space than we now give to windows.50
PHOTOMETRY.
The search for a primary standard of light still goes on, and
the requirements for one have been the subject of papers ap-
pearing from time to time during the past year. But little actual
progress in this direction has, however, been made public.
At the last meeting of the German Illuminating Engineering
Society, however, a paper was read discussing the question of an
absolute standard, and it was proposed to investigate the dis-
tribution of energy in the spectrum of the source of light to
be investigated, to determine the sensibility curve of the human
eye for different wave lengths, and to calculate therefrom the
distribution of light intensity in the spectrum, and finally to
arriye at the total candle-power. An extended discussion of the
possibilities of using the black body as the basis of a rational
unit of light followed the reading of this paper.51
The past year has been noteworthy in regard to the paucity
of published reports on secondary standards. In the National
Physical Laboratory of England a set of twenty-four tungsten
substandards running at 1.5 watts per candle has now been
established on a satisfactory basis. The values of candle-power
which are now assigned to these lamps have been determined by
a "cascade" method by six different observers, the work having
been in progress for about three years. In the cascade method it
was found that the mechanical errors and those due to changes
of observers from day to day have given rise to a probable
error of 0.08 per cent. The possession of a series of sub-
standards graded in the hue of the light emitted has been found
a valuable asset in the testing of lamps of different types in
60 Lighting Journal, February, 1913.
51 Journal fur Gasbel., July 12, 1913.
33^ TRANSACTIONS I. E- S. — PART II
enhancing the accuracy and facilitating the testing of such light
sources as acetylene lamps, gas mantles, flame arcs and metallic
filament lamps of all kinds.52
Tests were also made upon helium tubes, which indicated that
the maximum deviation in the tubes tested, and which were
selected with care as to uniformity of bore and thickness of wall,
was 3 per cent. The chief improvement now to be sought is com-
plete freedom from striations.
In phases of photometry other than in regard to standards, the
work during the past year has been on greatly varied lines. Con-
veniences in working the present type of photometer, selenium
and other photo-electric cells, the reliability of the flicker pho-
tometer, and the application of the radiometer have all been the
subject of study.
The photo-electric cell particularly has received much atten-
tion. Of great importance is the development of a new form by
the pioneer investigators of this cell, in which the alkali metal
surface is covered by a layer of hydride. This is made permanent
by filling the glass bulb with a rarefied atmosphere of argon or
helium. These special cells possess extremely great sensibility.
Recent experiments undertaken to determine the relationship
between the photo-electric current and the intensity of illumination
have shown a direct proportionality over a range of illumination
from about one-third that due to full sunlight down to the lowest
illumination detectable by the eye.
Other experiments have had for object the explanation and
elimination of a number of disturbing factors variously called
dark currents and contact electromotive force effects. It now
appears probable that in the photo-electric cell, when completely
developed, we shall have an objective photometer of very satis-
factory character.53
The selenium cell has been developed to such a point that it
gives excellent results with monochromatic light, provided the
exposure is short. An accurate sensibility curve has, moreover,
been established from which it was found that the cell was most
sensitive to yellowish-green light under faint illumination, and to
red light under strong illumination. The relations are, however,
62 Electrician, June 27, 1913.
53 Physik. Zeilsch., August 15, 1913.
REPORT OF THE COMMITTEE ON PROGRESS 337
so complicated that, except for monochromatic light the selenium
cell is as yet impracticable for photometric work.54
The efficiency of a selenium preparation used as a detector of
light may be defined as the amount of additional conductivity
imparted to it by the unit of incident light. Later investigations
show that the total effect of the light action is shown to be pro-
portional to the square root of the incident energy, while the
instantaneous effect is proportional to the energy. This is veri-
fied down to an illumination of o.ooooi meter-candle. It is
shown that selenium is the most efficient light detector known,
and is capable of discriminating minute differences of luminous
intensity far beyond the capacity of the eye.55' 56> 57
Color photometry has of late occupied much attention. A
chromoscope was devised a couple of years ago in which the
color could be determined in terms of the colors obtained from
the standard quartz plates by polarized light. In a newer instru-
ment four numbers are required to designate the color under
test as to tone, saturation and brightness. The quartz plates are
in two systems, but a wide range of color measurement can be
made using one system only. The author of the article describ-
ing the instrument considers it of the greatest advantage that a
color can be reproduced at any time from its four characteristic
numbers alone.58
A series of experiments made to ascertain the form of optical
instrument most suitable for color measurement showed that any
color could be imitated by the admixture in suitable quantities of
the lights of three suitably chosen narrow regions of the spec-
trum. Instead of adding these colored lights, broader primaries
may be subtracted from the white light by the selective absorp-
tion of suitably chosen dyes, but in this case the difficulty of
expression becomes serious owing to the impossibility of securing
dyes which will absorb one primary region and transmit the
remainder of the white light in equal proportions.59
About a year ago a simple method of comparing the colors of
64 Electrician, December 6, 1912.
55 Electrician, July 11, 1913.
66 Proceedings Royal Society, August 19, 1913.
57 Elec. Rev. and West. Elect., February 1, 1913.
58 Ann, der Physik., October 15, 1912.
59 Proc. Phys. Soc, October, 1912.
33§ TRANSACTIONS I. E. S. — PART II
artificial illuminants was described, based on the observation of
their comparative intensities through a photometer. In the eye-
piece of the instrument blue, green, yellow, red and deep-red
glasses are inserted successively, and in this way comparisons
were made of the spectrum intensities of these colors.60
In connection with color photometry, the flicker photometer has
received its share of attention and a discussion has taken place
as to whether this type does actually measure relative light inten-
sities. It has already been pretty well established that it gave
more consistent results when used to measure lights of dissimilar
colors, but there was a question as to whether there was not a
"luminosity sense" distinct from the "color sense."61 Recent
careful investigations indicate, however, that the flicker pho-
tometer gives true brightness, although it gives at low intensities
a reversed Purkinje effect which makes it necessary to use cau-
tion in its employment.62
It has been suggested that the physiological process, which
affords the common basis by which colored and uncolored lights
are measured by means of the flicker phenomenon, is the con-
traction of the iris when exposed to bright light. According to
this the flicker adjustment would be complete when the iris has no
tendency to alter under alternating illumination. An experiment
was made in which the irises of the observers were paralyzed
with a solution of homatropine sulphate. The appearance of the
flicker was, under these conditions, just the same as when seen
by normal eyes. All degrees of flicker remained as before the
atropine was placed in the eyes, and it was noted that the dis-
agreeable quality of the coarse flicker persisted undiminished.
Moreover the numerical results obtained for each eye and for
each observer agree very well among themselves.63
One writer claiming to have a practical solution of the problem
of heterochromatic photometry describes absorbing liquids, defi-
nitely specifiable, which may be used in varying thicknesses and
proportions to make the light of a given standard like that of any
other illuminant. A yellow and a blue solution have been found
which suffice to match all the ordinary illuminants with a Carcel ;
60 Good Lighting, August, 1912.
61 Philos Mag., November and December. 1912.
62 Electrical World, April 19, 1913.
63 Philos Mag., July, 1913.
REPORT OF THE COMMITTEE ON PROGRESS 339
a purple solution is suggested for use where these are not suffi-
cient, the three absorptions giving by the three-color principle
all the tints as represented in a color triangle.64
A suggestion made to eliminate color difference in the pho-
tometry of incandescent lamps is to run the standard lamp at
such a voltage that its color corresponds to that of the lamp
under test, and then determine the candle-power of the standard
from a previously plotted curve of candle-powers and voltages.65
To the number of illuminometers already on the market has
been added still another. This consists of a box containing a
small standardized tungsten lamp at one end and a bunsen screen
at the other. On one side of the box is a sliding rheostat con-
nected in series with the standard lamp and provided with a
pointer indicating on a scale the illumination on the screen in
foot-candles, the adjustment being made entirely by varying the
resistance. Two scales are provided, one reading from 0.3 to
10 foot-candles, the other from 0.00 1 to 0.3 foot-candle, cor-
responding with the application of 4 and 2 volts respectively to
the standard lamp.66
Still another recent illumination photometer consists of a
Lummer-Brodhun arrangement with lateral windows to which
are attached two rectangular tubes, one directed toward the screen
upon which the illumination is received and which may be set at
any angle and in any plane, and the other terminating in an
elbow leading to a standard tungsten lamp, in which elbow is set
a standard reflecting screen ; in this photometer the standard
tungsten lamp is movable, traveling along a scale.67
Another photometer, depending upon the acuteness-of-vision,
comprises a white wedge illuminated by the light to be tested.
Between the eye and the illuminated area is placed a glass screen
on which a diagram of very fine lines is marked out. In com-
paring two lamps the observer merely moves the photometer back
and forth until the lines become perceptible and notes the distance
from the source. In order to prevent imagination from assist-
ing the eye, the lines may be rotated so that the observer does not
know their position. The method is said to give satisfactory
M Transactions Illuminating Engineering Society, June, 1913.
88 Electral Review (London), November 15, 1912.
•* Illuminating Engineer (London), December, 1912.
07 Illuminating Engineer (London), September, 1912.
340 TRANSACTIONS I. E. S. — PART II
results, but allowance has to be made for the effect of adaptation
on the eye.68
During the year attention has continued to be given to pho-
tometers adapted to the newer requirements of the incandescent
lamp industry. Two watt-per-candle photometers have been
described having for their object determination of the voltage at
which the lamp under test will give the desired efficiency rather
than the candle-power, as was the former practice. One of these
is noteworthy in that the operation of the instrument is exactly
parallel to the operation of the formerly widely used candle-
power photometer as used upon single circuit fluctuating volt-
age.69
Another watt-per-candle meter has been devised based upon
the principle that if two potential differences are equal as indi-
cated by a galvanometer, then the two lamps are at the same
watts-per-candle. If they are different the variation in a resist-
ance, required to make them equal, is a measure of the ratio of
the watts-per-candle of the two lamps.
A rather novel method of reducing the quantity of light for
photometric purposes was described in a paper read before the
German Illuminating Engineering Society. If the light used for
illuminating the surface under test be reduced by placing in its
path an opaque disk with an open sector, the illumination of the
surface will not be uniform if the disk is stationary. The author,
therefore, lets the light passing through the sector be received
in an Ulbricht sphere, the inner surface of which is uniformly
lighted by reflection as far as it is not illuminated by incident
light. To complete the arrangement so as to have a portable
photometer for white light, a white ring surface, with an opening
in it is provided on the outside of the sphere. The observer
changes the angle of the sector until the hole in the ring becomes
invisible, thus obtaining a measure of the light by which the
white surface with the hole is illuminated.70
A novel method has been devised to measure the energy of the
ultra-violet radiation emitted by a mercury arc lamp. This is
based on the fact that the coefficient of velocity of hydrolysis of
68 Zeit. Instrumenlenk., September 1912.
69 Lighting Journal, July, 1913.
to Tnl.fur Gasbel., July 12, 1913.
REPORT OF THE COMMITTEE ON PROGRESS 341
tetrachloroplatinic acid varies proportionately to the incident rad-
iant energy.71
Investigators are still working on the application of photog-
raphy to photometry and it is expected that an emulsion will
be obtained which gives results comparable to the impression
upon the eye. Means have now been developed for measuring
intensities, opacities and other properties which it is necessary
to know of the developed emulsion, and the different materials
possible to use in the emulsion have also been investigated.72' 73> 74
An instrument has been devised for observing intrinsic bril-
liancy. This consists of a lens equipped with a "cat's eye"
diaphragm throwing an image of the source to be examined on a
screen inside a blackened box. By means of a Lummer-Brodhun
prism, the observer can compare the brightness of the image with
that of a diffusing plate illuminated by a small standard incan-
descent lamp. The adjustment of equality of brightness is
made by means of the diaphragm.75
OPHTHALMOLOGY.
Many articles have appeared lately upon the requirements
for hygienic lighting and the advantages of one system of light-
ing over another; these are largely, however, either recapitula-
tions of previous knowledge or are inspired by the advocates of
the method of illumination in question.
An investigation was made last year in regard to defects of
vision among school children in Liverpool, from which it ap-
peared that, as might be expected, vision was better among child-
ren attending schools in the outskirts where the open surround-
ings gave opportunity for open spaces than among those attending
schools in the center of the city. With one exception the girls'
sight was distinctly worse than the boys' this being attributed
by investigation to the fact that the girls' course included sewing.
This was regarded as the most likely of any item in the school
curriculum to affect the eyes, and is often carried out under the
most disadvantageous conditions of lighting. One investigator
71 Comptes rendus, January 27, 1913.
75 Le Genie Civil, February 15, 1913.
78 Comptes rendus, February 3, 1913.
'* Elek. u. masch.. May 25, 1913.
'* Comptes rendus, April 31, 1913.
342 TRANSACTIONS I. E- S. PART II
thinks it would be desirable if sewing could be abolished altogeth-
er under the age of seven, and further urges that there is not
enough discrimination during sewing lessons between children
whose eyesight has been found satisfactory and those with ser-
ious defects.76
Similar investigations have been carried on in other coun-
tries. While it has been shown that conditions producing defec-
tive vision are now much more closely studied than heretofore,
and that defective conditions are being remedied, the percentage
of children suffering from poor sight is still large. Such sub-
jects as size and style of type, quality and color of paper, illus-
trations, angle, position of reading, and color of blackboard and
chalk have all been, and are still being studied, particularly with
reference to saving the sight of the children.77
A series of experiments was made to determine the perception
of lights of short duration, the assistance of seventeen observers
of different ages and occupations being enlisted. From the results
obtained a curve was plotted showing the perceptibility of flashes
of light in terms of the durability and intensity.78 Tests were
also made by other experimenters along the same line.79
An interesting investigation was recently carried on by a
French journal to find the combination of colors most legible at
a distance. The order of merit is somewhat surprising, the best
being black on yellow, while the customary combination of black
on white appears sixth in a list of thirteen.80
A somewhat similar test was made to find, if possible, the
most legible type to use for books and periodicals. The type
proving best was the one selected by our newest illuminating
engineering journal, while a rather startling conclusion was
that the American typewriter stood in a class by itself as the
worst.81
Two experimenters have recently published the outlines of
recent researches lasting over a year upon the effects of radia-
tion upon the eye, most of the tests being made upon rabbits. The
T6 Good Lighting, September, 1912.
" Ophthal. Record, February. 1913.
rs Transactions Illuminating Engineering Society, November, 1912.
» Electrical World, August 9, 1912.
so Scientific American Supplement, February 15, 1913.
31 Lighting Journal , January, 1913.
REPORT OF THE COMMITTEE ON PROGRESS 343
light sources chiefly used were the quartz-tube mercury arc lamp
and the magnetite lamp. Much of the information obtained is
not yet in shape for publication, but it was shown plainly that
under the ordinary commercial conditions surrounding the use
of even the more brilliant illuminating agencies, no specific dan-
ger exists ; and that only by the grossest neglect and deliberate
protracted exposure of the eye to the brilliant light sources at
close range is there the slightest chance of injury to the organs
of vision, except in so far as temporary injury may be due to
the effect of ordinary visible radiation.82
Tests using commercial light sources and commercial colors
were also made to determine the effect of a colored background
upon visual perception. The results were not fully reported, but
they showed that the intensity of reflected light from an object
is a much more important factor in perceptibility than color.83
Investigation of the perception of color shows that the du-
plicity theory can no longer be held, but that all facts harmonize
with the theory that the rods are the organs which are concerned
in the perception of colors of short wave lengths.84
It has frequently been said that with natural lighting out of
doors, a higher intensity of illumination is necessary than with
artificial lighting. A series of tests were made at sunrise and
sunset on a number of observers, from which it was shown
that daylight illumination in the open is satisfactory for reading
at all intensities as low as with artificial light, and probably
lower.85
ILLUMINATING ENGINEERING SOCIETIES.
Since the last meeting of the Illuminating Engineering Society,
another national illuminating engineering society has been
formed: — the Deutsche Beleuchtungstechnische Gesellschaft.
This is the third association of the sort in existence, these beins:,
in order of time of formation, the American, British and Ger-
man. The first general meeting of the new society was held
in the physical laboratory of the University of Berlin on Feb.
24, 1913. Among the papers presented were: "The Eye and II-
82 Electrical World, May 24, 1913.
83 General Electrical Review, April, 1913.
84 Arch, oj Ophthal., March, 1913.
85 Electrical II '01 Id, July 5, 1913.
344 TRANSACTIONS I. £. S. — PART II
luminating Engineering," "Light Units," "The Boiling Tempera-
ture of Carbon in the Arc Lamp," "Method of Diminishing the
Light Intensities in Photometry," "A New Method of Deter-
mining the True Temperature of Filaments in the Incandescent
Lamp," and "The Present Disadvantages of the Sphere of Auto-
mobile Lighting."86. 87> 88
A gratifying announcement is that of the reorganization of
the International Photometrical Commission. This was organ-
ized by the International Gas Congress in 1900, and held its
first meeting in Zurich in 1903. This was composed of represen-
tatives of gas companies with the co-operation of certain of the
national laboratories. It is now proposed to extend the com-
mission to include not only the gas interests but also the electrical
interests, and to be representative of these industries, illuminating
engineering societies and other associations interested in photo-
metry and illumination and to be responsible to them. A sub-
committee has already been appointed to consider photometrical
units and standards.89'90
The British Illuminating Engineering Society has partly com-
pleted a study of school and library lighting with a view to
formulating recommendations suitable for general adoption.
Committees of this society, in co-operation with committees of
the Association of Teachers in Technical Institutions and the
Library Association have issued preliminary reports on the arti-
ficial lighting of schools and libraries calling attention to a
number of important features frequently overlooked, prominent
among which is glare. Recommendations as to minimum illumi-
nation for various purposes were made.91
RELATIONS OF ILLUMINATING ENGINEERNG TO OTHER
BRANCHES.
Illuminating engineering is gradually coming into recognition
by other professions and arts. The architect is beginning to
appreciate the value of a knowledge of the principles of illumi-
nation in decoration and utility, as is shown by the co-operation
86 Journal of Gas Lighting, December 10, 1912.
87 Electrical World, April 26, 1913.
88 Elek. Zeit., June 26, 1913.
89 Journal of Gas Lighting, December 17, 1912.
90 American Gas Light Journal, November 11, 1912.
91 Electrician, July 25, 1913.
REPORT OF THE COMMITTEE ON PROGRESS 345
of the architect and illuminating engineer in planning the light-
ing features of the coming Panama-Pacific Exposition in San
Francisco, and in designing the illumination of many important
buildings, such as libraries, churches, restaurants, etc. At the
International Congress of School Hygiene just held in Buffalo,
two sessions were completely turned over to the Illuminating
Engineering Society and presided over by a representative of
that society. The theatres are also coming to the illuminating
engineers for their scenic effects ; and in one installation of
especial interest, where the services of an illuminating engineer
were enlisted, gauze scenery and dyes were used in preference to
canvas and paint, and instead of using dipped lamps the effects
were produced by color screens. The lighting arrangement is
such that the lighting of the scenery is entirely separate from
that of the actors so that while the scenery passes through a
very large range of color changes, the color of the light on the
actors remains constant. Furthermore complementary spots are
used to correct the color of the actors' faces when colored light
has to be used on the stage proper. In this installation the use
of the arc lamp has been entirely abandoned and the concen-
trated filament form of tungsten lamp used entirely. Foot-
lights have been largely given up and the lighting of the stage
is arranged to produce natural effects by lighting from overhead
by front lights in the gallery and by lights directed from one side
of the stage so as to produce normal shadow effects. It is in-
teresting to note that these changes reduced the maximum load
from 45 to 5 kw.
Manufacturers are taking up the question of lighting their
shops and mills, and have investigated the influence of the
character of the light on the time taken to perform mechanical
operations. One of our large lamp concerns recently sent a
letter of inquiry to a number of industrial concerns; of the 209
replying, 164 said that improvements had been made in their
lighting, and a large number of expressions of opinions were
received to the effect that increased .production, better goods,
and greater satisfaction on the part of the workers had been the
result. Moreover, in many cases the cost of lighting has been
reduced.92
92 Journal of Gas Lighting, February II. 1913.
346 TRANSACTIONS I. E. S. — PART II
Merchants have been employing the illuminating engineer to
such an extent that it is rare that a periodical devoted to the light-
ing industry does not contain a description of the lighting fea-
tures of some large store in which the installation was carefully
planned and well carried out. This applies also to most estab-
lishments depending for their success upon the general public.
That the work of the railway mail clerk is one of the most
trying occupations as far as the eyes are concerned is now recog-
nized, and the Post Office Department has issued specifications
for lighting the mail cars, covering location of light source with
regard to the initial illuminating values, absorption by globes
and reflectors, light failures, emergency lighting, etc.93 The
data for these specifications were obtained from actual tests on
a mail car of one of our large railroads.94
In this connection it may be well to mention that an exhaustive
test has been made at the shops of one of our large railway sys-
tems to standardize the best method possible of train lighting.
This test was made under the auspices of the Association of
Railway Electrical Engineers.
The relation of lighting to the number of accidents has also
been receiving attention as never before. The intensity of the
illumination and the location of the sources have been studied
in connection with moving machinery, passage-ways and stairs ;
and in this connection it has been suggested that too much light
might be as bad as too little in the dazzling effect on the eyes.
A very practical illustration of the relations of light to acci-
dents is the statement of a leading casualty company of New
York that "the greatest number of accidents occur during the
months of diminished light." Furthermore, a prominent official
of one of America's largest manufacturing companies is author-
ity for the statement that "insufficient illumination" is frequently
held by juries to be "contributory negligence," and in the defense
of accident suits the lawyers of this company find it a valuable
point to offer testimony by a competent witness to prove the
adequacy of lighting arrangements in this company's plants.95
In this connection it may be said that several states have passed
ss Lighting Journal, February, 1913. -
94 Electrical World, March 29, 1913.
w Safety, p. S6.
REPORT OF THE COMMITTEE OX PROGRESS 347
laws specifying the candle-power of locomotive headlights,96 and
a number of tests, the last under control of one of the state public
utility commissions, were made to find out whether the lights
complied with the law and also to determine whether the head-
light might endanger the safety of operation of trains through
interference with signal lights or in any other way. With elec-
tric headlights, numerous phantom lights were seen, but these
were of the nature of mere flashes where the engine was running
at the ordinary speed. There was difficulty in distinguishing
classification lights and engine numbers on locomotives equipped
with arc headlights, and precautions should be taken to place these
marks at such a distance from the headlight that the latter may
not materially interfere with their correct reading. The arc
lamp should be switched off when passing through large yards
and other places where it might have a tendency to interfere
with the performing of duty by yardmen or others, or to en-
danger their lives. An incandescent lamp should then be
switched on.97
A report, rather startling in its novelty, has been issued to the
effect that souring of milk is due to ultra-violet rays, and that
milk may be kept sweet for days by putting it in a red glass
bottle, or in a plain glass bottle wrapped in red paper.98 A re-
verse action apparently on this order, and now creating much
comment, is the use of ultra-violet rays for sterilizing water,
as noted above.
A suggestion is contained in an installation of gas arc lamps
for lighting tennis courts. Perfect satisfaction was obtained, and
the courts were even better patronized at night when it was cool
than in the heat of the day.99 This installation paralleled a sim-
ilar one in another town in which tungsten lamps were
used,100' 101 and these, in turn, were followed by a golf putting
course lighted by gas. The latest item of this sort is the equip-
ment of a polo-field which is to be lighted by forty-eight 400-
96 Lighting Journal. January, 1913.
•7 Scientific American Supplement, February 1, 1913.
98 Scientific American, June 7, 1913.
99 Good Lighting, September, 1912.
100 Electrical ll'otld, July 12, 1913.
101 Electrical World, August 24. 1913.
34-8 TRANSACTIONS I. E. S. — PART II
watt tungsten lamps with metal reflectors, the installation being
designed to give one foot-candle on the horizontal plane.102
INDIRECT AND SEMI-INDIRECT LIGHTING.
Indirect lighting and its half-brother, semi-indirect lighting, are
progressing steadily, and numerous installations of each are illus-
trated in the current technical journals. Manufacturers are
almost daily producing new forms of fixtures of this type, some
of them artistic in the extreme. One fixture for semi-indirect
lighting has a space between the bowl and upper reflector en-
closed with clear glass to keep out dirt, insects, etc. This has
features of value, but its artistic merit is open to question.
As a relief from the present standard, though undesirable,
method of lighting railway cars, it may be noted that certain
express trains now running out of New York are now provided
with indirect fixtures.103
STREET LIGHTING.
The question of street lighting is attracting extraordinary atten-
tion. People are no longer satisfied with just enough light to see
to move around safely, but are coming to a realizing sense of the
advertising and artistic values of ample light. White Ways are
almost as common as towns themselves, and where the cities do
not seem inclined to install them, the merchants put them in by
private subscription. The open arc lamp is a back number, and
in many cases the enclosed arc, once so universal, is giving place
to magnetite or flaming arcs, or to clusters of tungsten lamps on
ornamental poles. The old, severely plain iron or wooden pole
is rapidly giving way to the artistic post, and there is a strong
tendency to recognize the artistic as well as the utilitarian.
While much publicity is being secured to ornamental lighting,
the majority of the business, however, has been connected with
the ordinary street lighting, which is, in the electric field, prac-
tically working in the direction of the use of the luminous arc
lamp, and the tungsten filament incandescent lamp. The increased
standard of general street lighting has been largely accelerated
by the reduced cost of light, and while it has taken on no dis-
102 Lighting Journal, May, 1913.
103 Electrical World, April 26, 1913.
REPORT OF THE COMMITTEE ON PROGRESS 349
tinctively new form, the standard of lumination intensity (not
necessarily size of units) has been raised very considerably.
An interesting test was made in Switzerland to find the rela-
tive advantages of arc and metallic-filament incandescent electric
lamps. Two streets of equal length were lighted with 10-ampere
electric lamps and 500-candle incandescent lamps respectively.
The choice between the two forms of lighting was left to 29
trolley car motormen. Of these 25 favored the metallic-filament
lamp on account of lessened glare and irritation to the eyes.104
The mercury-vapor lamp has been suggested as a street lighting
unit, but thus far its use for this purpose has been hardly more
than a suggestion, although in one city lighted by a municipal
electric plant, a group of merchants installed six quartz tube
lamps to show by contrast the poor character of the general street
lighting.
During a "street show" in one of our large cities, ornamental
pillars each carrying three electric light globes were erected.
Panels in these columns were made transparent by making them
of wire netting and coating them with varnish of various colors,
giving the effect, when lighted from the inside, of art glass.105
In the ornamental lighting system, it is frequently desirable
to turn off the lamps without affecting the rest of the circuit.
This is done in at least three towns by various systems of pilot
wires and magnet switches centering at the central station or
other convenient point.106- 107> 108 In this connection may be noted
a method of controlling from the central station switches on a
network by superimposing a ripple on the regular voltage.109 In
another town a switchboard has been placed in the office of the
chief of police so that in case of burglary or fire alarm after the
regular time of shutting off, the ornamental system of street light-
ing may be turned on.110
An unfortunate dispute has arisen in one of the large cities of
England over the relative merits of high pressure gas and flaming
arc lamps. Two streets were lighted by the rival illuminants and
IM Electrical World, April 26, 1913.
105 Electrical IVorld, October 12, 1912.
100 Electrical World, November 30, 1912.
107 Electrical World, September 7, 191 2.
108 Electrical World, September 14, 1912.
104 Electrician, February 14, 1913.
110 Electrical World, October 12, 1912.
350 TRANSACTIONS I. £. S. — PART II
experts representing each of the industries made illumination
measurements and prepared reports. The tests were entirely in
favor of the electric lamps, but the fact was brought out that the
gas lamps were improperly adjusted and installed, so that no con-
clusions could be drawn as to the relative merits of the two
systems. The whole affair caused much argument and a good
deal of acrimonious discussion on the part of the advocates of
the two systems.111- 112 0
Gas street lighting has made great strides in both England and
continental Europe, where high pressure lighting is in great favor.
Automatic lighting of gas lamps is also making rapid progress
on the other side of the Atlantic, the lamps in a large number of
towns being equipped with these appliances. Two systems of
automatic lighters are in extensive use, one being operated by a
momentary addition to the street main pressure and the other by
means of a clock arrangement on each post so that the lamps
operate individually and independently. Highly encouraging
reports as to the satisfactory and economical working of these
systems have been made.
In this country street lighting by gas, while making steady
progress, has not experienced the rapid growth that is so marked
abroad. Automatic lighting has not as yet obtained a foothold
here, and except for two minor installations, there is no high-
pressure street lighting. Another difference between European
and American practise is that abroad inverted burners are becom-
ing the universal practise, while here, except for a two-mantle
150 candle-power lamp, the inverted burner is not used.
Ornamental street lighting by gas is spreading, a number of
prominent installations having been made.
It may be of interest to note that one of our public utility
commissions, which had been investigating the street lighting in
a large city of the state, recommended that the city employ an
illuminating engineer, to be retained permanently if possible, for
the purpose of selecting the type of lamps to be used and to fix
their location.113
Increased interest in street lighting does not seem to be unusual,
111 Journal of Gas Lighting, October 5, 1912.
112 Electrician, March 7 and 14, 1913.
113 Electrical World, January 18, 1913.
REPORT OF THE COMMITTEE ON PROGRESS 35 1
however, as a recent item in a French technical journal contains
the statement that in that country, out of 10,000 villages or com-
munes of more than 1,000 inhabitants, 6,000 are without public
lighting.114
FIXTURES, GLOBES AND REFLECTORS.
Architects and decorators are realizing more and more the
importance of artistic and appropriate gas and electric light-
ing fixtures, and new designs adapted to all conditions are daily
appearing on the market. One type that is especially popular
just now is the "shower" chandelier, and designs of great beauty
have been brought out. Another new design is an indirect light-
ing portable which may be used as an ornamental table light.115
Still another fixture designed to give a strong concentrated light
for fine work consists of a small reflector socket carrying a
tungsten lamp and welded to substantial brass tube bent so as to
fit closely the body of the machine in connection with which it
is to be used. This tube is securely fastened to the machine
and becomes practically part of it.116
A novel fixture has been installed in the rotunda of one of our
state buildings, and is probably the largest chandelier ever built.
The fixture body is over sixteen feet high and is suspended by
a chain seventy-two feet long, consisting of twelve-foot links
containing special tubular tungsten lamps to give the effect of
a string of light, the joints in the chain being provided with ball
lamps in decorative design. The fixture itself is finished in
composition silver leaf. In some of the reading rooms of the
same building indirect lighting has been used with the bottoms
of the basins made of pink Georgia marble, thus producing very
beautiful effects. The general illumination of these rooms is
low, and so each table is provided with reading lights designed
by actually placing an individual at the table with a book and
adjusting the lamp for his comfort.
In another installation in an office building an attempt was
made to reproduce artificial daylight through a false skylight
matching the color received through an actual skylight in an
adjoining room. The lighting is so arranged that at night, when
114 L'Electricien, January n, 1913.
1,5 Electrical Revinu and (Vest. Elect, March 1, 1913.
116 Electrical World, November 2, 1912.
352 TRANSACTIONS I. E. S. — PART II
artificial light is used, the color of the illumination will be ad-
justed to match the artificial lighting.
In connection with the use of marble for diffusing material
noted above, it is reported that patents have been taken out in
Germany for using marble instead of glass. Marble is planed
down until it became translucent and different intensities of
light were shown from behind. The effect obtained was that the
illumination was hardly distinguishable from daylight, and it was
difficult to realize that the room was artificially lighted.117
Under this head might be mentioned efforts recently made to
increase the brilliancy of moving pictures. The available sources
of light having about reached the limit of their intensity, the
reflecting power of the screens is now receiving attention. The
early muslin screens were replaced with canvas coated with a
layer of white; ground glass, which was next tried, was found
too fragile, and was in turn replaced by a fabric coated with
aluminum powder giving a screen presenting a silvery surface
of great uniformity. The new screen is 3.7 times as luminous
as the muslin.118
An investigator conducted a series of tests to determine the
distribution of light in an ordinary room. Under the conditions
of the test it was found that the light was strongest in the upper
and lower part of the room and less intense throughout the
middle portion. Working on this idea, a translucent screen was
made in the form of a vertical half cylinder placed close to the
wall ; between it and the wall was set the light source. Meas-
urement of the light distribution under these conditions showed
that it was almost identical with that obtained from daylight.119
About the first of last year a screen was brought out which'
filtered the rays of the electric arc light in such a proportion
that those rays passing through it formed a true daylight color.
This was followed a short time ago by a similar screen for the
incandescent gas lamp. The resulting light is a perfect match
for average daylight, and by it colors may be judged with perfect
accuracy.120
i" Good Lighting, January, 1913.
118 Good Lighting, January, 1913.
n9 Transactions Illuminating Engineering Society, June, 1913.
120 Lighting Journal, May, 1913.
REPORT OF THE COMMITTEE ON PROGRESS 353
PHYSICS.
Previous determinations of the constant of the Stefan-Boltz-
mann law of radiation vary from 5.3 to 6.5. During the past
year, two independent investigations were made from which new
figures for the constant were derived.1-1- 122
Another investigator describes experiments showing the devia-
tion from Lambert's cosine law of tungsten and carbon at glow-
ing temperatures. It was found that the brightness of tungsten,
beginning with normal emission, increases with the angle of
emission, reaches a maximum at about 750, and for larger angles
diminished rapidly. The brightness of carbon, beginning with
normal emission, decreases with increasing angle of emission,
the rate of decrease increasing with the angle. The relative
brightness variations for tungsten at the higher of the two tem-
peratures chosen are about 20 or 25 per cent, greater than the
corresponding variations for the lower temperature. No defi-
nite change was found for carbon.123
An experimental lecture was delivered late last spring before
a European society on the relations between spectral analysis
and the electronic theory. The fundamental point was that
what oscillates in light is nothing but electrons. The lecturer
also discussed optical resonance as deduced from the theory
of electrons and the Zeeman phenomenon.124
LEGISLATION.
An appreciation of the importance of proper illumination is
growing on the legislative bodies of different countries, and
investigations are being held and laws passed governing particu-
larly the lighting of factories and workshops. In Holland, the
law specifies that the employment of women and young children
is forbidden in works in which artificial light is normally re-
quired between 9 a. m. and 3 p. m. An illumination of 1^ foot-
candles is specified as the minimum for certain processes ex-
ceptionally trying to the eyes, and a minimum of one foot-candle
in less exacting occupations.125
121 Electrician, December 20, 1912.
122 Deutsch Phys. Gesell., November 15, 1912.
123 Astrophys., December, 1912.
i24 Elek. Zeil., April 23, 1913.
154 Iron Age, August 17, 1912.
354 TRANSACTIONS I. E. S. — PART II
In England, a special committee has been appointed by the
Home Secretary to inquire into conditions necessary for ade-
quate and suitable lighting of factories, and in France the ques-
tion has been under discussion for about two years.126' 127
In our own country the New York State Factory Investigating
Commission has been studying the lighting of workshops for two
years,128- 129> 130 and a bill has been drafted with the aid of a
committee from this Society for its regulation. The Industrial
Commission of Wisconsin last January issued its general order
on sanitation which included ventilation and shop lighting. This
order provides for daylight illumination and specifies the minimum
illumination under different conditions.131
PHOTOGRAPHY IN ILLUMINATING ENGINEERING.
An interesting lecture was delivered not long ago illustrating
the value of photography in illuminating engineering. The
speaker laid stress on the difficulty of obtaining good photographs
by artificial light, and said that there was little information avail-
able as to what the exposure ought to be or how to allow for
the actinic values of the different kinds of light. The two essen-
tials in a good photograph of an installation are that the room
shall appear exactly as it really is by artificial light, and that the
positions and natural appearance of the fixtures shall be shown
without halation or distortion. The lecturer believes that the
ideal way to preserve a good record of lighting installations is
to have a really good photograph showing the actual installa-
tion as it appeared to the eye and also data on intensity of
illumination.132
In another lecture on photography by invisible light, the
speaker said that the longest infra-red rays thus far measured
(those of the quartz) have a wave-length of 0.3 mm., while the
shortest electrical waves observed are 2 mm. in length, indicating
a brief undiscovered gap between the two sets of phenomena.
The lecturer showed ultra-violet photographs of the invisible
126 Journal of Gas Lighting, January 2S, 1913.
127 Electricia?i, January 24, 1913-
128 Gas World, December 14, 1912.
I** Journal of Gas Lighting, December 17, 1912.
wo Electrical World, December 2S, 1912.
131 Gas Age, March 15, 1913.
is-' Illuminating Engineer (London), December, 1912.
REPORT OF THE COMMITTEE ON PROGRESS 355
electronic discharge which proceeds from the ordinary electric
arc and not detectible to the eye. A current of air diminishes
the intensity of this discharge within its own range, but does not
affect the streamers beyond. The ultra-violet photographs pre-
sented were most interesting in showing the diffusion of the ultra-
violet shadows even in full sunshine. Ordinary glass is prac-
tically opaque to light of this short wave-length, while the pig-
ment "Chinese white" appears black under its illumination. The
lecturer's lunar photographs also show hitherto invisible patches
of heterogeneous material near one of the craters indicating
strongly the possibility of sulphur deposits and so contributing
to the evidence of their volcanic origin. The infra-red landscape
views were remarkable for their black skies and strong shadows
and for the snowy whiteness with which the green foliage
appears, owing to the deep-red component of its chlorophyl color-
ing matter.133
ILLUMINATION MEASUREMENTS AND CALCULATIONS.
Some study has been put upon methods of determining and
calculating illumination under various conditions, and a number
of ways of shortening and simplifying existing means have been
put forth. One writer suggests that if a curve be plotted in which
the y-axis, corresponding to the relation of the intensities of the
lights, be divided to a logarithmic scale, and the x-axis, corre-
sponding to the distance from one of the lights to the screen, be
divided to a natural scale, the resulting curve will have a much
more useful character than if drawn on ordinary graph paper.134
A method has been worked out for determining the illumina-
tion at any point on a flat surface illuminated by a source above
it. The method consists in dividing up the surface into areas
each corresponding with a unit solid angle. By means of the
formulae and tables given, the illumination can be calculated more
exactly than by the approximate methods generally employed.135
One author criticises the present method of laying out the light
distribution curves of any source, and reepmmends an illumina-
tion distribution curve in which the lengths of the polar ordinates
are proportional to the product of the intensities into the areas
138 Electrical World, February 8, 1913.
184 Archiv.fur EUklrotechnik, 1, 1912.
la5 Eleklrolech. Zeitsch., December 19, 1912.
356 TRANSACTIONS I. E- S. — PART II
of the zones in which the rays are taken. The same writer calls
attention to the fact that an error is introduced in the illumination
measurements by neglecting the more or less efficient utilization of
the reflected light according to the angle of emission from the
light source.136
Respectfully submitted,
F. N. Morton, Chairman,
F. E. Cady,
E. L. Elliott,
G. L. Hunter,
S. G. Rhodes,
Frank E. Wallis,
W. R. Burrows,
Dr. F. Park Lewis,
T. J. Litee,
Bassett Jones, Jr.
Committee.
DISCUSSION.
Prof. George A. HoadlEy: — After hearing this report I am
more than ever convinced that I have not fully appreciated the
progress of illuminating engineering. I have just one remark
to make in regard to photography for establishing standards of
light, and what I have to suggest, I suppose has already been
done. I speak of it simply because I am not sure. Those of us
who are accustomed to practical work in photography know the
tremendous difference there is in the results obtained dependent
upon the condition of the illumination of the subject, and upon
the length of time of exposure. If we are going to have any-*
thing at all that will give us standards along the line of actual
illumination through photography, a standard time of exposure
and condition of illumination must be agreed upon.
136 Progressive Age, December 2, 1912.
lewis: value of light, shade, form and color 357
THE PSYCHIC VALUES OF LIGHT, SHADE, FORM
AND COLOR.
BY* F. PARK LEWIS, M. D.
Synopsis: In its last analysis the physics of light must he considered
in the effect of shade, form and color upon the human emotions, feelings
and sensations. No single factor more definitely dominates the lives of
men than the impressions made upon them by what they see. The dig-
nity and beauty of our surroundings inspire to higher living and better
citizenship. We are brought in relationship to the external world only
through our special senses. Were all other senses abolished while the
intelligence remained, there would be no possibility of communication
with the outside world. Light and color are not the waves of different
amplitudes in the ether but are the results of the impressions produced
upon our consciousness by these external influences. We perceive objects
only by reason of the reflected light from their surfaces. Could we
imagine a condition in which surfaces would reflect no light, objects
could not be seen even though light were present. The element of beauty
has not only an esthetic value; there are by-products, if they may be so
called, which are incidentally developed and which have even a greater
bearing upon our lives. In a mining district in Pennsylvania the work-
men were encouraged by competitive prizes to make beautiful gardens in
back-yards which had been filled with debris. The effect was not only to
beautify the district and to give an agreeable change of occupation with
physical betterment, but an increase of civic interest. The mental and
moral effect of light and shade cannot be ignored. Excessive lighting
brings out sordid details with unpleasant glare which is as bad in effect
as insufficient light. The beauty of light sources should not be forgotten.
Good lights for the poor will make the home beautiful. Collaboration is
urged on the part of illuminating engineers, architects, school men,
ophthalmologists, and others to secure broader instruction on the care of
the eyes.
It would seem a far cry from the physics of illumination and
its application to the affairs of civilized life to a study in
psychology in one of its most subtle and illusive phases, and a
word of apology and perhaps of explanation may be necessary
in justification of my temerity in asking a group of practical
scientists who are dealing with real problems and actual things
to go so far afield from the noise and bustle of the street as to
penetrate into the mysteries of the mind itself. I think you
will agree with me however that the subject is not so remote
nor abstract as might at first sight appear.
358 TRANSACTIONS I. E. S. — PART II
It makes very little difference how we may analyze the various
spectra in determining their ultimate construction, or with what
care we may study the nature of the light sources and the phe-
nomena of the reflection and distribution of the luminous energy,
if we fail to take into account as a primary and essential consid-
eration the effect, not only upon the human eyes, but upon the
human emotions, feelings and sensations, produced by the changes
of light and shadow, of complementary and contrasted colors and
the influence actually produced on our actions and conduct by
the things which constantly come within the range of cur vision.
There is probably no single factor which more definitely dom-
inates our lives than the impressions made upon us by the things
which we see. It makes a vital difference in his outlook on life
and his attitude towards society if a man's surroundings are sor-
did and dirty and mean. Under such circumstances it will be
hard for him to be a good citizen. What matters to him wheth-
er or not we have dignified and inspiring architectural monu-
ments if he lives in a home but little better than a
pig-pen? How can we hope that he will be interested
in good government, in public art galleries, open air
music, or in well lighted streets, if he goes from digging a
sewer into a hovel scarcely more attractive? It is just the
uplifting glimpses of beauty that take him out of the drudgery
of a monotonous existence and make him realize that his own
home may in simple ways be made more livable and more at-
tractive. While this is in no sense a paper on social welfare the
far reaching effect of light and shade of form and color is so
great as to excuse the somewhat unusual form in which the
subject is presented.
We know of course as a scientific fact that the only way in
which we are brought into relationship with the things sur-
rounding us is through the medium of our special senses, that
these are the pathways through which we are made conscious
of the existence of the external world. If one of these senses
is lacking or deficient we must learn to depend upon the others.
While there does not exist an automatic or compensatory bal-
ance by which one sense is increased in efficiency by reason of
the loss of another the very necessity of depending upon those
LEWIS : VALUE OF LIGHT, SHADE, FORM AND COLOR 359
which remain to us may, and often does, make them sharper and
more quickly responsive. An almost completely blind lad whom
I saw a few days ago was readily able on hearing coins jingled in
the pocket to give the number of them and their denomination.
This is not surprising when we recall that each piece of metal has
a distinctive tone when struck against another and that it required
only an alert and correct ear to differentiate the tones and to as-
sign to each its proper value. The degree of accuracy possible
when certain senses are trained was brought very vividly to my
attention sometime since, when I had the unusual opportunity of
spending a few hours with that rarely gifted woman, Helen Kel-
ler. Her hearing and sight were lost at so early a period in her
life as to exist only as the vaguest memory. Her only remaining
avenues of communication with the outside world are through
her sense of touch, of taste, and of smell. The sense of taste
is not highly cultivated with any of us and the opportunities in
which it can be used as a method of discrimination are relatively
limited and infrequent. There remain then for her only the
sense of touch which has been refined to such a degree as to
make it interpret to her volumes that to the rest of us are
closed ; and the sense of smell, which while not as keen as that of
a hound, is so vastly finer than that which most of us possess
that it serves as a reliable aid in a large number of circumstances
to enable her to locate herself, to determine who are her com-
panions, and in a word to bring to her mind a multitude of
facts for which most of us depend upon the employment of our
other senses. The thought occurred to me that if it were pos-
sible to conceive of the existence of a still further loss of per-
ception so that there would be no method of conveying to her
brain the vibrations that are carried through the floor upon which
she stands, or through the trunk of a tree by which she recog-
nizes the swaying of the branches and the moving of the leaves,
— and if it were still further possible to imagine the loss of the
sense of smell so that the odors which are carried on the air
would convey nothing to her intelligence, and if with this were to
go the sense of taste and it were still possible that all her func-
tions could be carried on, there could remain immured within the
prison walls of her body the same intelligent, responsive, percep-
360 TRANSACTIONS I. t,. S. — PART II
tive, even intutive intelligence that now exists; but there would
be absolutely no method by which she could be brought in con-
scious communication with the outside existing world. Indeed so
far as she is concerned there would be no zvorld because its exis-
tence is predicated upon its recognition by the intelligence and
sensation within. The impressions which are conveyed to us
therefore by our senses are tangible and real only to the degree
that they are recognized and understood.
It is a very old subject of discussion as to whether sound would
be produced if a bell were rung out at sea where there were no
ears to hear it; or if in the absence of eyes the trees would still
be green and the poppies red or the rainbow, when right condi-
tions exist, still in the sky. Of course there is neither sound,
nor light, nor form, nor color in the absence of an intelligent
recognition of these qualities because it is not the motion in
the air produced by the pounding of the hammer upon the bell
which constitutes sound; it is the impact of these atmospheric
waves upon the tympanum carrying an impression through
sentient nerves to special brain centers and the transformation
of these impressions into a conscious recognition of that which
we understand as the sound produced by the ringing of the
bell. The colors of the rainbow are not the vibrations in the
ether of wave-lengths of the different amplitudes, but the result
of the impressions which these vibrations produce upon the
retina, exciting sensations in the rods and cones which are con-
veyed through the optic nerves to the visual area and are there
interpreted into our conscious understanding of that which is
known to be the red, the yellow, the orange, the blue, the green,
the indigo and the violet.
In a recent discussion before the Oxford Ophthalmological
Congress on "Nystagmus" or twitching of the eyes of min-
ers, which is produced by working in semi-darkness, the follow-
ing interesting hypothetical experiment was proposed.
Imagine that you are in a cavern, the floor, walls and roof of which
are absolutely devoid of color, and having surfaces of such a nature that
they reflect no light; imagine that they are covered with some substance as
lamp-black, only much blacker.
Put a lighted candle above your head, so that the light may fall in
every direction, but not into your eyes. What can you see? Nothing
but dead black.
lewis: value of light, shade, form and color 361
Double your illumination, have two candles, and you will see "dead
black." Then take a 20 candle-power lamp instead of the candles, and
you will see nothing more than "dead black."
Try a searchlight, train it on the wall opposite, and imagine that a
large black beetle, that has covered itself with the lamp-black-like sub-
stance, is crawling up the wall in front of you. You will not be able to
see it.
It is impossible to distinguish one piece of "dead black" from another.
It is evident that under such conditions increasing the light does not
assist vision ; it is simply wasting light.
Light, in the absence of all color, and in the absence of all reflections
from surfaces, is useless for vision.
Now, imagine that you are in a brilliantly colored chamber — no matter
how bright the colors may be — but that there is no light. What can you
see? Again, nothing.
Color, therefore, in the absence of light, does not exist.
Now imagine that you are back in the cavern again, which is devoid
of color but that the surfaces are crystalline, such as coal or jet, that
is to say, they have innumerable small reflecting surfaces, or "surface
brightness." With one lighted candle above your head, you will be con-
scious of black surfaces with innumerable specks of glistening white light.
This is the light that is reflected back directly into your eyes without
diffusion, and represents one element in the "surface brightness" of the
wall, and it can be measured.
Double the brightness of your sources, and you will get about double
the amount of this surface brightness. Have four times the light and you
get four times the brightness.
Let the beetle, now with a clean and glossy back, crawl up the wall
again. You will be able to see a white spot of light moving, but you will
not be able to recognize that it is a beetle
The conclusions to be drawn from this feat of imagination
are : ( 1 ) that light, without color or surface brightness, is
useless for vision; (2) that color, without light, does not exist;
(3) that light without color, with surface brightness enables
you to see a little.
In this connection it is interesting to consider what are the
processes by which we become conscious of any existing object.
Let us imagine what would take place in the brain of a child
to whom any object, an orange, for example, was presented for
the first time.
Every impression comes as a new one. The eyes are open but
the images which have been carried to them have been vague and
indistinct. They have not been differentiated. The sound of
voices comes as a murmur or a noise, possibly broken by varia-
362 TRANSACTIONS I. E. S. — PART II
tions in intensity, as from the crash of a falling body, or the
jangling of a bell; but the nicer discriminations of sound, which
may be developed with increasing refinement until the most ex-
act harmonies or the least discordant note is at once recognized,
have not yet begun. The first impression conveyed to the eyes
of the child after making the distinction between light and dark-
ness would be that of color. When the orange was brought in
his range of sight, its brilliancy would attract his notice and he
would become conscious of a blotch of color, like the sunshine
or the lamp-light which he has already seen, and this, in the
sight center situated at the back of his brain, would occasion a
flood of nervous energy, excited by the vibrations in the ether,
and the neurons of the terminal nerve endings would respond
with a quickened capacity to apprehend, to appreciate this same
phenomenon when it again occurred. It would be the beginning
of those finer color distinctions that zvere to come later and
which were to constitute the education of his color sense, which
zvere to enable him to understand, to appreciate and to feel the
beauties of the color harmonies of the world in which he lives.
Then gradually this splotch of color, otherwise so meaningless,
would take on form, and he would realize that it was limited by
a circle, and in gaining this knowledge a new group of cells
would be energized, and another essential fact in relation to his
surroundings would have been achieved. Then would come to
him the realization of a new and more wonderful phenomenon:
the circle has depth; and here an enormous advance has been
made. He has been introduced into three dimensional space.
Both sides of his brain are working synchronously. He has
binocular vision. The image which has been made upon the
retina of each eye has been carried to corresponding parts of
the brain, overlapping and blending as in a stereoscope. A mul-
titude of new impressions have been aroused, suggestions of the
outer world, indeed of the universe, have been conveyed to him.
His logical faculties have been awakened, and without either
knowing or realizing it he has done the most important thing
in the world. He has begun to think. The thought which he is
unable to express has aroused his will, has excited his desire,
has tempted him to experiment. He timidly and tentatively
reaches out his hand and touches the thins: that he has seen and
LEWIS : VALUE OF LIGHT, SHADE, FORM AND COLOR 363
in recognizing the consciousness of its presence he has done the
most wonderful thing in the world. He has established the exis-
tence of a problem, the solution of which has been the basis of the
speculations or our most profound philosophers, from Plato to
Kant. He has located himself in space. He has begun to find him-
self. He has commenced his education. He will now learn to dif-
ferentiate between soft and hard, between rough and
smooth, between elevation and depression, between those
things which oppose and those which attract. With each
new idea has come a new flood of energy, sweeping through
his brain and making the pathway easier for those which
are to follow. The skin of the orange is broken and
the fragrance of the volatile oil is carried to his nose. It
brings an odor like nothing else in the world, yet it is one of
hundreds of perfumes and scents and smells that he is to learn
to differentiate from all others, and with sensation comes thought,
with thought, suggestion, and with suggestion will, and the motor
influences which are to govern this will during all of his life have
been established.
The orange drops from the baby hands and falls with a dull
thump to the floor. It gives a sound that to the trained ear con-
veys intelligence of the nature of the thing itself. It is at once
evident that that which has fallen is not metallic; it has not
the flat sound of a closed book nor the overtones of a hollow
wooden box. It is one of a thousand possible sounds and yet
it carries a descriptive story to the listening ear and trained
brain. Finally the fruit is retrieved, the skin is removed, the
segments are broken apart, and everyone of these movements
with the little sounds connected with them are educative. When
at last a portion of the fruit is conveyed to the mouth they
arouse, — who can say how many groups of motor influences :
the whole body is moving, the neck bends, the arms, the wrist,
the fingers, each with its separate centers in the brain represented
by neurons, almost without number. When the segment is
placed in the mouth the muscles of the lips, the tongue, the jaws,
the cheeks — all are dominated by impulses which are sent out
from the brain — and finally when the juice of the fruit is tasted
and this wonderful complex of sensations has been united into
what has been termed a sterognostic comprehension of the whole,
364 TRANSACTIONS I. E. S. — PART II
the entire brain has been excited into activity from front to
back and from side to side giving instructions in co-ordination,
in will, in logic, and perhaps in ethics. Each group of nerve
centers that has been energized in receiving impressions or in
sending out commands is being educated in the only way in
which it can be to perform the work which it is ultimately des-
tined to do.
So that not only the perception of form or color is an essen-
tially psychic process but by the automatic relationships which
are aroused through what are termed the association fibers in
the brain, other emotions are excited, and what would seem to
be a simple, becomes a most complicated process. The thing
that we. see may give rise to emotions far removed from that
which might be naturally expected. For example, a landscape
of most extraordinary beauty may be associated with some
earlier circumstance of a repellent character and not only will
that particular view be ever afterward disagreeable but things
associated with that view of which we may be quite unconscious
when they appear in other places and under other conditions
may excite that same unpleasing sensation. It shows conse-
quently that there can be no real or exact psychic values, for not
only are the things which we see modified in our recognition of
them by things which we have seen, frequently very early in
life, but they may be so modified also by associated contigencies
that there can be no exact and invariable value which is not
modified by previous impressions. In consequence of this fact
every view which meets the eyes is a composite of the thing seen
with an additional modifying element which may be in some
cases absolutely changed by the supplemental impressions which
we bring to it. This has been so definitely recognized by the
modern school of artists that they have long since realized that
the more exact the reproduction of a landscape or of a face the
less like is it to the original. It is impossible for any artist
to place in his picture the fugitive impressions which are rapidly
chasing each other over the features, or the flash in the eye
which so illuminates the character, or the droop of the lip which
may betray a weakness, and all of which are instantaneous.
Monet painted seventeen views of a haystack because
as the day changed, it was never twice alike. But if
LEWIS : VALUE OF UGHT, SHADE, FORM AND COLOR 365
he can suggest in vaguest outline that which he feels and sees
he has made it possible for the understanding observer to supple-
ment in his own mind that which has been suggested. The
indefinite, hazy, shadowy, landscape then becomes vitalized by
the associated memories which it has stimulated. With the
definite, sharp-cut, precise reproduction of the actual form of
the thing itself staring him in the face, as in a photographic
representation of it, all of these subtler but therefore more real
qualities which differentiate that thing from all other things in
the world are masked. A mask not only conceals, it distracts
the mind from that which is beneath it and prevents that play
of the mind which enables one to build out of the shadowy sem-
blance all of those beauties with which we would wish to see it
invested.
All men of imagination, whether they are artists in stone or
in words, whether they are the discoverers in science, the lead-
ers in finance or the makers of an empire, are essentially and
potentially poets and the poet is he who short-circuits truth
through the fourth dimension of intuition.
We are all influenced emotionally by form and color. We
unconsciously feel the depression produced by the black gar-
ments of widowhood, and the enlivening cheering effect of.
brightness and color. We feel the color atmosphere in which we
live. They who live in a land of clouds reflect in their charac-
ters and bearing the shadows that fall upon them ; while the
sunshine glows in the lives, in the mentality, and in the activi-
ties, of those of southern lands. As the heavier and more mas-
sive forms of architecture have a morally depressing effect,
as dark walls and sombre furniture drink in the light, as dim
rooms and badly illuminated corridors are forbidding and fore-
boding, so is the converse true. So, too, the dazzling glare of
brilliant lights brings out in actual detail all the sordid fittings of
a poor room or the inharmonious settings of a badly furnished
one. We feel instantly the atmosphere of the place we enter.
The importance of beauty as an essential element in civic better-
ment and in its social aspects has not received the attention
which it deserves. If we are influenced to the degree which I
have indicated by our environment it must follow invariably
that surroundings which give an atmosphere of quiet and which
366 TRANSACTIONS I. E. S. — PART II
pleasurably excite the imagination must have a beneficial moral
effect upon the community. The atmosphere which exists to-
day is one of excitement and the tendency is to increase this
excitement by all forms of artificial stimulant. Our newspapers
not satisfied with four-inch headlines in order to attract
notice have adopted green and pink outer covers as an added ap-
peal to the eye.
THE UTILITY OF ART.
The element of beauty in our common life has not only an
esthetic value which in itself is important but there are certain
by-products, if they might be so-called, which are incidentally
developed and have even a greater bearing upon our lives.
In a mining district within an hour's journey from Pittsburgh
a few years ago the homes of the miners were sordid and un-
kempt, the streets being littered and the back yards repositories for
all sorts of rubbish. The corporation in charge of the community
introduced in the management of the plant improvements in ac-
cordance with the most advanced methods, and among other
measures prizes were offered for the best garden plots to be
found about the homes of the workmen. This has resulted in an
eager and intelligent competition and in place of an offense to
the eye the town has become a beauty spot. Fences are lined
with rows of hollyhocks and golden rod, the walks are bordered
by attractive arrangements of garden flowers beautifully kept;
vegetable gardens have become productive and in some instances
have added to the annual income of the miners as much as $100.
Miners working in dark coal shafts straining their eyes
to see and having them dazzled by millions of reflections from
the shiny surfaces suffer from an affection known as miners
nystagmus, the rapid oscillation of the eyes, to which reference
has already been made. I am told that the refreshing change
from the gloom of the mine to the soft colors of the
garden in which they work has already exercised a most bene-
ficial influence upon this serious affection. In its moral effect in
giving an appreciation of the values of better civic conditions in
developing a civic spirit, interest has been aroused ; they have not
only made gardens they have made men and citizens.
The mental and moral effect of light and shadow, the difference
LEWIS : VALUE OF LIGHT, SHADE, FORM AND COLOR 367
produced upon our state of mind by the glaring brilliancy of an
unshaded Welsbach light, especially an old one, or the soft glow
of an even yellow illumination is felt by every one although by
no means always recognized as a cause of nervous irritation. In
some of the most persistent cases of eyestrain after the ophthal-
mologist has employed the highest degree of skill in determining
the correct combination of lenses to be employed it will be found
that the discomfort is due to a badly placed lamp, to the improper
use or absence of shades, to an insufficiency or an excess of light,
to some specular reflection, or other local fault in the illumina-
tion about which he has not been advised.
There is probably no one simple element that more deeply con-
cerns the welfare of all people than correct lighting.
In the studies in efficiency in lighting we seem to have forgotten
the beauty of light itself. No mere luminosity will replace a
visible light source. By grouping lights of low power and prop-
erly planning them — using translucent, frosted, or prismatic
globes, effects of great beauty may be secured. It would seem
unnecessary to put emphasis too strongly on the superiority of
indirect illumination.
It is in the dark streets and unlighted alleys that crime skulks.
Good lights are cheaper than policemen and more effective.
When the dark corners are lighted we are ashamed to have
them cumbered with debris and we clean them up.
When a man comes home after a hard day's work, stumbles
through a dark hallway and finds the living room dimly lighted
by an unshaded lamp with a smoky chimney, glaring, yet in-
sufficient, bringing out all of the misery without softening any
of its harsher outlines, is it to be wondered that he makes his
stay as short as possible, seeking in preference the brightly lighted
saloon where appeal is made to his eyes as well as to his appetite ?
Until we make the homes of the poor fit habitations for them to
live in we cannot expect them to spend much time in them, nor
can we expect to make good citizens out of them. How can we
hope for civic pride or civic righteousness to come out of an
unlovely dirty tenement house. A stream will not rise higher
than its source, and the source is the home and the family.
One of the easiest, one of the least expensive methods of mak-
368 TRANSACTIONS I. E. S. — PART II
ing the poor home livable would be to introduce good lights in it.
Could a more effective or a more helpful propaganda be inaug-
urated than to teach the poor to light their homes adequately,
beautifully and cheaply? This could be done with a minimum of
effort, for many of the homes are lighted imperfectly, in an ugly
way and at an extravagant cost. It should be one of the first steps
in the new movement for the conservation of vision.
It ought not to be a difficult matter to secure proper lighting
for our public institutions. Of all buildings those which should
demand good lighting are our public libraries. Still to-day when
so much has been accomplished on these lines, it is the exception
rather than the rule to find a public library in which the lighting
is not atrocious.
IMPORTANCE OF GOOD LIGHTING.
The importance of good lighting in public buildings is so self-
evident that it would not require mention were it not for the fact
that often in the finest specimens of modern architecture this
seems to have been overlooked.
Some time since I happened to be in one of the progressive
western cities where the State house, a splendidly located edifice
on a hill, which was notable for its beautiful approaches, was so
poorly lighted that on entering the relatively small doorway on a
clear sunshiny morning it was found that the entire main floor
was artificially illuminated and the basement floor in which were
situated some of the most important offices for the transaction
of the business of the state could not have been used were it not
for the artificial lights employed. Unhappily this is not an ex-
ceptional circumstance, alike serious defect is found in the multi-
million dollar State Capitol at Albany, N. Y.
In our auditoriums the lights, to paraphrase the meaning of
the apt French expression, "jump to the eye." In unnumbered
public schools to-day, in which artificial lights must be used, the
children are facing flickering gas lights in a vain attempt to see
the marks on shiny black-boards. The school authorities have
not yet learned that dark red and green walls absorb the light for
which the children are suffering.
The time has fully arrived when an authoritative body com-
posed of architects, of illuminating engineers, of school-men, of
VALUE OF LIGHT, SHADE, FORM AND COLOR 369
ophthalmologists and of all others who have to do with the man-
agement of light or the use of the eyes should collaborate in the
development of authoritative plans for the education of the public
on sight protection. The American Medical Association is now
forming a sub-committee from the medical societies in every state
in the Union on the conservation of vision. The National Edu-
cation Associaton is deeply interested and is now ready to sup-
port any proper effort for broader instruction on the care of the
eyes.
It was proposed several years ago that there might be one
day in the year given to the conservation of sight. It would be
of great interest for the children to study the condition of their
own schools. In this respect their essays might include the
physics of light, illumination, natural and artificial, the amount
of window space necessary for a well lighted room, how it should
be placed, in a word the hygiene of the eyes. The study which
this would necessitate would give a groundwork of knowledge
which would promote better conditions in the future than exist at
present. It would constitute a practical lesson on one of the es-
sentials of right living and would result in collateral benefits of
inestimable importance.
DISCUSSION.
Mr. G. H. Stickney : I can not pretend to discuss this paper.
On the other hand, I feel that we must express a special appre-
ciation for it. It seems to me that it is one of the most valuable
papers that I have listened to in a very long time. To those of
us especially who work largely from the engineering end it brings
a point of view which should modify our thought, and round out
our practise.
Dr. H. E. Ives: Mr. President, I wish to second, if I may so
put it, the remarks of Mr. Stickney. I think that it would be
safe to say that we have never in our history had presented to
us so clearly and impressively the importance, the almost sub-
lime importance, of illuminating engineering. We have had
pointed out to us among other things the importance of associa-
tion. We are accustomed to a certain kind of lighting and are
apt to argue because we have been adapting to that kind of
370 TRANSACTIONS I. £. S. — PART II
lighting through the ages that it is necessarily best. But after
all much of it may be a question of association. I talked not long
ago with a prominent member of this Society, who told me that
the people of his house had become entirely adapted to another
system of illumination than the one to which they were formerly
accustomed although the new system at first seemed all wrong.
It opened my eyes to the fact that if the associations are properly
planned it may be that we can very easily improve upon daylight,
or anything else that you want to present as ideal. It gives us
an opportunity. For instance, we can light a room by light
from the side or light from overhead, and we can adapt our-
selves to either one. We can allow ourselves to be guided by
other considerations than first impressions. We can search out
new lighting effects.
Dr. Lewis' has spoken of the healthful effect of proper associa-
tions and proper lighting conditions. Has it ever occurred to
you that we might start a new school of medicine. We have all
sorts of "paths" who have been in a correspondence school for
six weeks and learned all there is to be learned about healing.
May we not look forward to the "photopath" who will put a
patient in a calm, restful room, and subject him to lighting ef-
fects to subdue or stimulate him until his nervous condition im-
proves ?
Another point which Dr. Lewis has made which I think is
of very material commercial importance is that the matter of
lighting is getting to be recognized as an aid to architecture.
Formerly an architect designed his building as it would appear
in daylight. Then the lighting was put in as a necessary evil —
and it usually was an evil. Now, due to the application of a great
many minds — and very artistic minds — to the lighting problem,
we see to-day lighting installations which are appropriate. So
that, speaking for myself, I would prefer very often to see the
room lighted up by night rather than to see it by day, simply
from the beauty of the light sources.
As I said before, we ought to be guided by the paper brought
before us by Dr. Lewis to realize the real solemnity of the sub-
ject that we are handling.
CLAUDE: NEON LIGHTING 371
NEON TUBE LIGHTING.*
BY GEORGES CLAUDE.
It is known that the discovery of the rare gases, —
those very curious gases which were contained in the atmosphere
unknown to us — is the magnificent work of Sir W. Ramsay. *It
is known also that it was the distillation of liquified air which
led Ramsay to the securing of such wonderful results, results
which are all the more marvelous since they were obtained with
a modest apparatus producing from 1 to 2 liters of liquified air
per hour.
I had imagined that by using the much greater facilities at my
disposal and with the use of apparatus which can liquify 10,000
cubic meters of air per day, I might have obtained some new
results. But alas, I found that there was nothing to do after
Ramsay.
But if in spite of my desires, I have been unable to add to the
list of rare gases, I have nevertheless been able to produce them,
especially so in the case of the neon, in far larger quantities.
Things are so arranged in my oxygen apparatus that this neon is
the residue of the progressive liquefaction of air, and is ob-
tained as a by-product of the industrial manufacturing of oxy-
gen. The output of this apparatus is so large, that in spite
of the insignificant proportion of neon in the air, 1 part to
66,000, yet with a modest apparatus of 50 cubic meters of oxygen
per hour, 100 liters of neon can be produced in a day. Balloons
can be filled with this gas as I am showing it to you here, and
balloons that can fly, as the density of neon is two-third times
that of air.
Consequently, neon being such an abundant industrial product,
I have engaged myself in a search for uses for it.
I have directed my efforts toward light production. I do not
have to tell you, gentlemen, what a serious drawback the ever
increasing, dazzling and blinding properties of modern lighting
possess. And you are all aware of the hopes that have been en-
* Outline of a lecture given at the seventh annual convention of the Illuminating
Engineering Society, Pittsburgh, Pa., September 22-26, 1913.
372 TRANSACTIONS I. E. S. — PART II
tertained for the uses of diffused lighting without any fatigue to
the eyesight and which might be secured from the marvelous and
fine brightness with which the rarefied atmosphere in the Geissler
tubes is illuminated. Unfortunately, these last have remained up
to now detestable apparatuses, since the luminescent properties of
ordinary gases are not very good. Nitrogen, is the only gas used,
thanks to the remarkable perseverance of Moore, although the
efficiency of his apparatus is very low — 1.7 to 2 watts per candle.
Rare gases are remarkable for their ability to become lumines-
cent; their spectra are remarkable; the one of neon is especially
so. It contains some numerous and superb lines of red, orange
and yellow, and three important lines of green. Unfortunately, it
contains neither white nor violet tints. Certainly, this absence of
blue is a big fault so far as a source of lighting is concerned ; but
I supposed it possible to correct this fault, and I have passed on
to something else.
I had in view other reasons, besides the richness, of its spec-
trum, to be taken up in a study of neon. First, neon really
possesses an extraordinary aptitude to become luminous. It is a
long time since the illustrious Sir. J. Dewar has succeeded
in the production of tubes with neon which are illuminated
spontaneously at the points of maximum amplitudes of Hertz'
interferences and which are extinguished at the nodes. So
that these peculiar detectors furnish very original means with
which to measure the length of the waves in the installations of
wireless telegraphy.
And it was another one of Ramsay's colleagues, Professor
Collie, who with the neon that I had sent to Ramsay, has been
able to make the following observation, a very important one in-
deed: with a sealed glass tube containing a little quicksilver in
a rarefied atmosphere of neon, if one shakes the tube in a dark
place, the mercury looks like a real rain of fire. This curious
phenomenon is explained by the result of Bouty's experiences.
This scientist has indeed been able to observe with great surprise
that neon is easily passed through by an electric discharge : where
it is necessary to have 1,000 volts, for instance, in the case of air,
but 13 volts are sufficient with neon; and you can then appre-
CLAUDE: NEON LIGHTING 373
ciate Collie's phenomenon: the electrifying of the tube by the
shaking of the quicksilver is enough to cause discharges in the
gas.
With such properties, neon can very well enter the field of
production of luminescence. I have therefore, directed my ef-
forts toward using this gas in immense Geissler tubes, similar to
the Moore tubes. The first obstacle encountered was indeed a
queer one. You know, gentlemen, that there is also another name
for rare gases — noble gases. Now, it would appear that neon en-
tertains a very lofty idea of its dignity; it is quite capable of
working wonders by itself; it absolutely refuses to perform any-
thing when it finds itself in contact with those inferior fellows
known for instance as hydrogen or nitrogen. When it is re-
membered that these gases are much more inferior than it, either
in their aptitude for luminescence or with respect to luminous
efficiency, even minute traces of them mixed with neon are suffi-
cient to displease it to such an extent that none of the lines of
its spectrum are visible in the light produced. Here is a tube
which has 99 per cent, of neon and 1 per cent, of nitrogen. Only
the light produced by the latter appears.
It is not enough, therefore, to introduce into the luminescent
tube absolutely pure neon, for if at the start the tube shines with
a magnificent brightness, the impurities emitted by the electrodes
when the current is passing through, cause very rapidly the drop
of its luminescence. In order to overcome this serious difficulty,
I was compelled to devise a process which would purify the neon
in the tube itself, as fast as the impurities were introduced into it
by the passage of the current. I was led with this object in view,
to make use of one of the curious resources of liquified air; the
remarkable property discovered by Dewar that charcoal absorbs
air with great energy when frozen to the temperature of liquified
air.
But it is under peculiar conditions that I use this property.
Charcoal does not mingle indifferently with every and all gases.
Generally speaking the harder they are to liquify the harder it is
to absorb them ; this absorption is very much smaller in the case of
neon than with the different sorts of gases which might accom-
pany it. And you can then conceive easily the process which I
374 TRANSACTIONS I. E. S. — PART II
have devised in order to purify the atmosphere of my neon tubes.
The tube to be rilled is connected to a charcoal receiver, immersed
in liquified air. This charcoal slowly absorbs the gases developed
through the passage of the current, but it leaves the neon. By
this contrivance a pump has been made; but this is an intelligent
pump which sucks and carries away the troublesome molecules,
and respects the others. After a laborious process which lasts
for many hours, the neon remains victorious ; the tube is finished,
sealed off from the charcoal receiver, and will show thereafter
without weakening the superb light of the neon. Superb ! well
all tastes are different and you will perhaps find that I exagger-
ate; but I shall try just now to give you a more complete satis-
faction.
This difficulty having been overcome, another one cropped up.
I had noticed that the neon tubes thus obtained were short-lived.
After showing a rapid increase in the difference of potential
at the bars, they began to flicker and to crackle, and lastly went
out in 5 or 6 hours. Well, gentlemen, you will easily recognize
there the strange phenomenon discovered by Moore. Moore has
observed indeed that the atmosphere inside of his tubes was rare-
fying itself progressively and that finally the tubes went out en-
tirely; this fact, a very mysterious one indeed, had neutralized
his efforts up to the moment when he had the idea of introducing
nitrogen in his tubes, by means of an ingenious electromagnetic
valve, as fast as the rarefaction took place.
Unfortunately, such a remedy could not be applied in the case
of neon for if Moore's observations are accurate, the quantity
of nitrogen absorbed by his tubes is astonishing: 200 liters per
year for a tube of 50 m. If my neon tubes were such gluttons,
my apparatus of liquified air would be inadequate to meet their
demand. It was of course necessary for me to look for condi-
tions which would permit me not to consider the absorption of
neon, so that one charge, one single dose of neon, could insure
to the tube a very long life, one comparable to that of an incan-
descent lamp.
To attain this result, it was necessary to begin by determining
the manner of this absorption. I was able to observe at the
outset that the electrodes of the first tubes, which were very
CLAUDE: NEON LIGHTING 375
small, became incandescent upon the passage of the current and
volatilized rapidly. A metallic deposit used to form from these
scales and strips in the neighborhood of the electrodes. I
have thought that it was this metallic deposit which on forming
absorbed the neon. And in fact, by dissolving these deposits in
nitric acid, gases were developed containing neon.
Therefore, there is no doubt that the volatilization of the
electrodes is what makes the trouble. In order to minimize it, the
use of large electrodes will be necessary henceforth, which can
be very little volatilized by the current. Experience has confirmed
this supposition and to such an extent that by using electrodes of
5 sq. dm. per ampere the volatilization is rendered nil and the life
of the tubes lengthened considerably.
The life increases naturally with the length of the tubes, as
there are always only two electrodes to absorb the neon, and that
the longer the tube the more neon it contains. With tubes 6 m.
long, you can attain easily 1,000 or 1,200 hours, and I have tubes
of 20 m. in excellent condition after 2,000 hours. This is super-
ior to the incandescent lamps.
Here is the problem solved then, and solved in a manner unex-
pectedly simple. We are in possession of tubes capable of show-
ing the spectrum of neon in all its purity. These tubes, which are
not provided with valves, are much more simple than those prev-
iously made, and give an entirely satisfactory length of life.
What advantages have these tubes as compared with the nitro-
gen tubes? Gentlemen, I shall only insist upon the essential
points and any one desiring further details is referred to the
paper which I read before the Societe des Electriciens, November
8, 1911.
At the outset the necessary difference of potentials is three
times less than with nitrogen ; this is a great advantage with re-
spect to safety. These 6 m. tubes have less than 800 volts at
their ends. Three could be mounted upon a transformer of 3,500
volts.
Secondly, its illuminating power is higher, 200 candles per
meter instead of 60. You can consequently use for lighting tubes
which are much shorter and accordingly less expensive. Another
3/6 TRANSACTIONS I. E- S. — PART II
advantage to be derived is that these tubes can be manufactured
in factories and carried all ready for use to a customer or client.
And lastly — this is the most important part — the illuminating
efficiency is much greater. Instead of 1.7 watt per candle ob-
tained by the use of nitrogen, only 0.5 is required for the long
neon tubes. May be you do not consider this marvelous when you
make a comparison with arc lights ; but it has to be borne in
mind that the question is one of spheric and not hemispheric
watts. Furthermore, no expensive carbons are required and all
upkeep charges are done away with. Really, when everything is
considered, I believe that with the exception of the mercury
lamps, the neon tubes supply the most economical lighting.
In some interesting experiments carried out at the Laboratoire
Central d'Electricite, Messrs. Broca and Laporte have observed
that the neon light is physiologically excellent on account of its
dull luminescence and that it increases visual acuity by 25
per cent. You can really notice with what clearness and sharp-
ness the small figures on the reports which I am passing around
can be seen.
All is rather perfect therefore but for the color, which is an-
other matter. Evidently this light is too red ; it is too red because
of its want of blue. Look at this bouquet ; it is of a beautiful blue ;
see how dull and disappointing it is. On the contrary here are
some poppies : see how resplendent they appear. No doubt, this
predominance of the red color allows of some beautiful illuminat-
ing effects. Here you have as an instance, the Grand Palais at
the Champs Elysees, in Paris, lighted by neon in 1910 on the
occasion of the automobile show held then; and the St. Ouen
Church of Rouen which was lighted by 50 neon tubes during the
festival of the Norman millenium. Undoubtedly, in a number
of cases of industrial lighting this light could be applied in
preference to that of mercury light, since it is very economical.
Allow me indeed to insist and to bring to your notice with what
strange facility you have accustomed yourselves to such a red
light, so as to retain only a pleasantly warm impression of a
golden yellow from which the red is completely absent.
But I have to admit that this excess of red is hardly acceptable
in the majority of cases and I have applied myself to having
CLAUDE: NEON LIGHTING 377
this light corrected. There is a solution which seems to me plain
and that is to combine the pale mercury light with that of the
bright neon. In this case, however, there are two difficulties to
be overcome. The first is that mercury and neon when placed in
the same tube will not blend and work together; second is that,
if it be required to mingle blue tubes with red tubes, the Cooper-
Hewitt mercury-vapor tubes would require a continuous current
at a low tension, whereas the neon tubes demand alternating cur-
rent at high tension. These do not go together. I have, however,
made correcting tubes, which are similar to the ordinary neon
tubes but having a little mercury. These tubes light with the
alternating current as in the case of usual neon tubes; but mer-
cury volatilizes progressively and the blue light which mercury
gives, invades the whole tube.
Well, here is progress, to be sure. Our blue bouquet has
recovered its colors, but it is our poppy bouquet which now looks
pitiable ; and as to ourselves, gentlemen, instead of being rubi-
cund, we are now just ghastly pale.
Patience! I light this neon tube and here we have the sun's
light succeeding to that of the moon's pale light. See how every-
thing has returned to its normal state; the blue color of this bou-
quet, the red of that one, the delicate hues of those flowers so
varied ; and above all you have noticed, ladies, how your complex-
ion matches nicely with this light. And what interesting orna-
mental effects can be obtained by the combination of tubes of
different colors. Here is an example, a fixture put up by the
firm of Paz & Silva for the automobile show in Paris.
The efficiency of these correcting mercury tubes is, unfortu-
nately, notably inferior to that of the neon tubes and should be
in the near neighborhood of i watt per candle. It is acceptable
even at that. Having an equal number of blue and red tubes,
you will see that a very pleasant light is obtained, very much dif-
fused, without shadows and at an energy consumption of 0.8
watt per candle. This is better than what has hitherto been
obtained with luminescent lighting.
Gentlemen, I have another application to bring to your notice.
If the objectionable red features in this light cannot be always
neutralized, there are cases where it proves to be an unquestion-
37& TRANSACTIONS I. E. S. — PART II
able advantage. To begin with, for illumination of monuments,
as I have already remarked; but it is of inestimable value for
advertising illumination, where the more dazzling the light is,
the more it will strike the eye, and hence the better it will be.
Now, with neon you are liberally served.
I have been able to make with the aid of my collaborator, M.
de Beaufort, some tubes of a small diameter which can be bent,
or twisted without difficulty and be given any desired form, and
be lighted with red or blue lights. There might be some appre-
hension that the minute quantity of neon contained in those tubes
might give them only a short life. However, I had the pleasant
surprise of finding out that with the sole condition respecting
the rule of the feeble density of the current at the electrodes,
these small tubes lasted as long as the big ones. Here is a tube
which has burned 1,400 hours. This one operates at the rate of
30 milli-amperes, sufficient as you can see, to give to it a very
luminous aspect.
These small apparatuses work on a common transformer with
alternating current. Upon continuous current, the transformer
is controlled by a rotating or Wehnelt interrupter. The cost of
an installation is not much higher than that of ordinary apparatus
and the consumption of energy is less and affords a much better
effect. There is certainly to be found here a brand new method
for the industrv of luminous advertising.
ROSE : LABORATORY OF GENERAL ELECTRIC CO. 3/9
THE ILLUMINATING ENGINEERING LABORATORY
OF THE GENERAL ELECTRIC COMPANY.*
BY S. L. E. ROSE.
Synopsis: About the year 1895 the General Electric Company started
at the Lynn works the study of illumination problems and the proper
application of arc lamps. In 1909, the department was moved from Lynn
to Schenectady. The present laboratory is equipped for making tests on
all kinds of illuminants. The work of the laboratory is divided into
four main divisions: namely; commercial investigations and their appli-
cations, photometric testing and developmental, research, and photo-
graphic. Facilities are available for a thorough investigation of all means
and methods of artificial illumination and the testing of lighting units for
commercial or special work. Parts of the laboratory equipment are
described and illustrated in this paper.
Before giving a description of this illuminating engineering
laboratory and the work being carried on there, perhaps a review
of its growth from its inception to its present proportions will
be of interest.
About the year 1895, the General Electric Company started
at its Lynn Works the study of illumination problems and the
proper application of arc lamps. This was the beginning of what
is now known as the illuminating engineering laboratory and the
work was carried on under the direction of Mr. W. D'A. Ryan
who is director of the present laboratory. The first photometer
room occupied a floor space of about 500 square feet (38.09
sq. m.). The photometer was of the single-mirror crane type
and could be used as a constant length, constant intensity or
constant radius photometer. All photometric testing was done
on this one photometer and it was a number of years before more
space was devoted to this work.
In the fall of 1907, three rooms, wi.th an aggregate floor space
of approximately 1,500 square feet (114.27 sq. m.), were built
in a new factory building and three photometers installed, one for
small units, one for large units and one for miscellaneous work.
*A paper read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
38o
TRANSACTIONS I. E. S. — PART II
In 1909 the illuminating engineering department was moved
from Lynn to Schenectady. None of the photometers was
moved from Lynn and the ones now in use at Schenectady were
designed, built and installed under the supervision of the depart-
ment. A description of the photometers used at Lynn and the
ones in use at Schenectady for large unit work as well as the
methods of test have been given in a previous paper before this
society.1
The present laboratory is situated in a two-story brick building
with an aggregate floor space of approximately 7,000 square feet
(533.27 sq. m.). (See Fig. 1.)
DARK / PHOTOGRAFIC
ROOM I DEPT.
SHOW
LECTURE
AND
DRAFTING M
ROOM
SHOW-ROOM
a
ULBRICHT
SPHERE
2M.0IA.
SECOND FLOOR PLAN
GENERAL
OFFICE
5 /<
RESEARCH AND
CALIBRATION ROOM
ULBRICHT
SPHERE
IM.DIA
AT
O'
\ <£■:
WORK ROOM
FIRST FLOOR PLAN
Fig. 1. — Plan of laboratory.
:oal )
Two universal rotators and a single-mirror crane type pho-
tometer (Fig. 2) are available for obtaining candle-power distri-
bution curves of small units. The latter photometer is very easily
changed into a constant length, constant intensity or constant
radius photometer. It is mostly used as a constant radius pho-
tometer and the radius of test-may be changed at will from 5 feet
(1.524 m.) to 20 feet (6.096 m.) or any intermediate point accord-
ing to the size of the unit being tested. For convenience of cal-
1 1. E. S. Trans., page 641, vol. 6, 1911.
ROSE: LABORATORY OF GENERAL ELECTRIC CO.
38l
culation, all tests, so far as possible, are made at a radius of 10
feet (3.04 m.). A sector disk is employed to increase the range
of the photometer, when necessary. For obtaining total flux
on small units, absorption of small units, globes or balls, an
Ulbright sphere of 1 meter diameter is used and for large units
an Ulbright sphere of 2 meters diameter is employed. In addi-
tion to the regular photometers mentioned above, a number of
portable photometers, of both foreign and domestic make, are
available for outside tests and for special work in the laboratory.
For spectrum analysis, color absorption, research and special
investigations, a spectrophotometer, spectrometer and colorimeter
are available.
Fig. 2. --Single-mirror crane type photometer for small light units.
Energy for lighting, power and experimental purposes is avail-
able from the direct current three-wire shop circuit of the General
Electric Company, the alternating current three-wire commercial
city circuit, a motor driven alternator which can be arranged to
give 25 to 60-cycle current, two motor driven direct current gene-
rators of 500 to 700 volts and 125 volts respectively, and a
constant current transformer for alternating current series work.
A switchboard panel is so arranged that current from any of the
above sources may be switched on any circuit in the building. In
addition to this, a circuit leads to each room from a 60-cell storage
382 TRANSACTIONS I. E. S. — PART II
battery. This battery is primarily for furnishing energy for
incandescent unit work and working standard lamps, but may be
used as an emergency supply, if necessary.
For the testing of gas units, the laboratory is equipped for both
high and low pressure work. For measuring consumption, wet
test meters are available which will operate on pressures from
2 to 3 inches (5.08 cm. to 7.62 cm.) of water up to 10 pounds
(4.53 kilograms) per square inch (6.45 sq. cm.). A motor driven
compressor is installed which will take gas from the city mains
and deliver it at pressures ranging from 2 or 3 inches (5.08 or
7.62 cm.) of water up to 10 pounds (4.53 kilograms) per square
inch (6.45 sq. cm.). Indicating and recording pressure gauges,
an indicating gravitometer and a recording calorimeter complete
the equipment. (Fig. 3.)
The work of the laboratory is divided into four main divisions,
namely : commercial investigations and applications, photometric
testing and developmental, research and photographic. The
whole comprises a thorough investigation of all means and meth-
ods of artificial illumination and testing of lighting units for
commercial or special work.
The commercial division is constantly giving advice and fur-
nishing lighting recommendations to all parts of the world and
for all classes of lighting of which the following may be men-
tioned as an illustration of the diversified character of the work ;
the office building of the Buffalo General Electric Company,2 the
Panama Canal, and the Panama-Pacific International Exposition
to be held in San Francisco in 1915. In connection with the
commercial division, a display of various types of lighting
units is maintained together with intensity and color booths
(Fig. 4). The latter two are probably the demonstrations of
greatest interest to the public. The terms used by the expert in
the art of illuminating engineering often seems, to the layman,
vague and inexpressive. The merchant may be told, by the engi-
neer, that he requires for proper illumination of different depart-
ments of his store, 2, 5, 7 or 10 foot-candles, but owing to the
vagueness of his conception of the foot-candle, he is still neces-
sarily "in the dark," as to the amount of illumination he is con-
tracting to buy.
2 For full description see I. E- S. Trans., vol. VII, page 597, 1912.
Fig. 3.— Gravitonieter and calorgraph.
3*
t
Fig. 4. — Demonstration room.
rose: laboratory of general electric co. 383
The idea of the intensity booths is to show the intensity of
illumination in steps of }4, Yz, l\ 3. 5» 7^> IO> :5 an(i 20 foot-
candles. This demonstration is made in a row of booths extend-
ing along one side of a room, each booth measuring approximately
2 ft. by 3 ft. 6 in. (0.61 X 1.07 m.). By the manipulation of
switches, these same booths are used to demonstrate the differ-
ence in color of the ordinary illuminants now in use and daylight.
The photometric division furnishes illumination data to all the
other divisions, outside departments, sales offices and through
them to the general public. Thorough tests are made on all kinds
of lamps and lighting equipment. Street lighting and interior
systems are tested under operating conditions. Experimental
tests are carried on night and day when necessary to try out
some new piece of apparatus for the patent department, or fur-
nish special data to the engineers of the laboratory. New designs
of lighting apparatus and systems are constantly being devised,
constructed and tried out to determine their commercial value or
their application to some special purpose. (See Fig. 6.)
The research division carries on special investigations of a
scientific nature.
The photographic division furnishes the commercial division
as well as the sales offices, architects, engineers, etc., with day
and night views of representative installations as well as many
conditions illustrating problems which are encountered and
examples of both good and bad lighting.
The purpose of the laboratory is utilitarian and altruistic and
it is devoted to the services of producers and consumers of arti-
ficial light and to the betterment of the art of illumination.
TRANSACTIONS
OF THE
Illuminating Engineering Society
Published monthly, except during July, August, and September, by the
ILLUMINATING ENGINEERING SOCIETY
General Offices: 29 West Thirty-Ninth Street. New York
Vol. VIII
NOVEMBER. 1913
No. 8
Council Notes.
A meeting of the Council was held in
the general offices of the society, 29
West 39th Street, New York, November
14, 1913. Those present were: C. O.
Bond, president; Joseph D. Israel, gen-
eral secretary; V. R. Lansingh, C. A.
Littlefield, L. B. Marks, treasurer;
Preston S. Millar, C. J. Russell, F. J.
Rutledge, W. J. Serrill and G. H.
Stickney.
The meeting was called to order at
10:30 a. M. by President C. O. Bond.
The minutes of the October meeting
were adopted.
Mr. C. A. Littlefield, chairman of the
Finance Committee, reported that his
committii had approved vouchers Nos.
1476 to 15 12 inclusive, aggregating
$1,646.25. Payment of these vouchers
was authorized. The committee also
recommended an appropriation of $100
as the quota of the society towards the
expenses of an International Commis-
sion on Illumination. The appropriation
was granted. Mr. Littlefield, also re-
ported that the Finance Committee
would submit at the December Council
meeting a budget for the present admin-
istration.
A report on the accounts of the society
for the year of January 1 to September
30, 1913, which had been prepared by a
certified public accountant, at the re-
quest of the Finance Committee of the
previous administration, was received.
The report will be published in the
Transactions. The earnings for that
period amounted to $8,256.49; while the
expenses, including an estimate of out-
standing debts, aggregated $8,175.65.
Mr. Israel reported that the total
membership of the society, counting the
resignations and applications presented
at the meeting was 1,392, and that the
expenditures for the first month of the
present fiscal year had aggregated
$2,893.17. The receipts during that
period amounted to $291.22.
The following amended report of the
Council Executive Committee, covering
business transacted by the committee
since the previous Council meeting, was
adopted :
A meeting of the Council Executive Com-
mittee was held in the general offices of the
society, October 31, 1913. Those present were
Chas. O. Bond, president; Joseph D. Israel,
general secretary; C. A. Littlefield, L. B.
Marks, treasurer, and Preston S. Millar.
Mr. Preston S. Millar was appointed chair-
man of a Committee on Education of the
Illuminating Engineering Society. The former
name of this committee was Committee on
Collegiate Education. It was suggested that
among other things the committee be asked
to consider the preparation of a tentative
course in illuminating engineering for schools
and colleges.
Mr. C. E. Clewell was appointed chairman
of a sub-committee on Office Lighting of the
Committee on Popular Lectures. An appro-
priation of $25 was granted for stenographic
expenses in connection with the work of the
sub-committee.
TRANSACTIONS I. E. S.— PART I
The following additional committee appoint-
ments were made: Finance: A. Hertz, W. J.
Serrill; Section Development: Joseph Langan;
Membership: C. J. Ramsburg; Board of Ex-
aminers: W. Cullen Morris, chairman, and
C. H. Sharp; Editing and Publication: Clay-
ton H. Sharp, chairman; A. S. McAllister,
VV. J. Serrill.
Twelve applicants were elected mem-
bers. Their names appear on another
page.
Ten resignations were accepted.
The following report pertaining to
the society's policy in supplying copies
of papers to authors was received and
adopted :
Your Committee, appointed to reconsider the
policy of the society in supplying authors
with copies of their papers which are pre-
sented before meetings of the society, begs
leave to submit the following recommendations:
When a paper is printed in advance, twenty-
five (25) copies shall be sent to the author.
When an author furnishes to the general
office the names and addresses of ten (10) or
less, non-members of the society, to whom he
desires to have copies of his paper sent, the
general office will, if copies are available, mail
to each of these persons a copy of the Trans-
actions containing the paper in question, indi-
cating at whose request it is sent.
That the standard form of reprint be
changed to secure minimum cost compatible
with good appearance; the pagination of the
paper, as printed in the Transactions, being
considered satisfactory. This will not pre-
clude the issuing of more expensive reprints
when so ordered.
That a suitable notice disclaiming responsi-
bility of the society for statements or opinions
of authors be printed on the title page of all
papers, whether in the Transactions, advance
copies or reprints.
Respectfully submitted,
Herbert E. Ives,
C. H. Sharp,
G. H. Stickney, Chairman.
The report was adopted and a vote of
thanks extended to the committee.
In accordance with a recommendation
in the foregoing report it was voted to
place the following statement on all
papers of the society :
The Illuminating Engineering Society is not
responsible for the statements and opinions
advanced by contributors.
Mr. V. R. Lansingh reported verbally
on the progress of the work of his Com-
mittee on Sustaining Afembership.
Reports on section activities during
the past month were received from the
following vice-presidents : Mr. G. H.
Stickney representing New York, Mr.
W. J. Serrill representing Philadelphia
and Mr. J. W. Cowles representing New
England.
Mr. Israel reported on the activities of
the Pittsburgh and Chicago Sections.
The following appointments to com-
mittees were confirmed : Nomenclature
and Standards, A. E. Kennelly, chair-
man; Research, H. E. Ives, chairman;
Advertising, R. E. Campbell; Papers,
E. J. Edwards, George S. Barrows,
H. A. Hornor, C. E. Stephens, M. G.
Lloyd, Alexander Duane; Reciprocal
Relations, W. J. Serrill, chairman;
Progress, T. J. Litle, Jr., E. L. Elliott,
T. W. Roth, W. E. Wickenden, H. S.
Hower, Wendell Reber ; Glare from
Reflecting Surfaces, F. A. Vaughn,
N. M. Black, J. R. Cravath, F. H. Gil-
pin, M. G. Lloyd; Lighting Legislation,
Ellice M. Alger, Oscar H. Fogg, Her-
bert E. Ives, Clarence L. Law, F. J.
Miller, G. H. Stickney, L. A. Tanzer,
W. H. Tolman.
It was resolved that the Council bf
the Illuminating Engineering Society
extend a vote of thanks to Mr. Joseph
B. Gregg for his valuable assistance in
arbitrating an account presented by the
Hill Publishing Company.
Informal mention was made of a re-
cent movement started in the West to
organize a San Francisco or a Pacific
Coast Section of the society.
The meeting was adjourned at 1 :i5
P.M.
TRANSACTIONS I. E. S.— PART I
Section Notes.
CHICAGO SECTION
A meeting of the Board of Managers
was held in the Grand Pacific Hotel,
November 5. Those present were : Dr.
M. G. Lloyd, chairman; J. B. Jackson,
secretary; J. R. Cravath, M. J. Sturm,
and H. B. Wheeler. After considerable
discussion of a program for the present
year, it was decided that it was not
desirable to arrange a definite program
for the full year, on account of the
arrangement of a number of joint meet-
ings with other societies, the dates of
which could not be determined at the
present. Mr. Sturm was delegated to
arrange a joint meeting with the
Chicago Architectural Business Men's
Association for January.
A meeting of the Chicago Section was
held in the Auditorium of the Western
Society of Engineers, Monadnock Block.
Chicago, November 12. A paper entitled
"The Illumination of Street Railway
Cars" was presented by Messrs. L. C.
Porter and V. L. Staley of the General
Electric Co. An excellent exhibit of
reflectors, holders, fittings and accesso-
ries was arranged by the Exhibition
Committee, of which Mr. H. B. Wheeler
is chairman. Mr. W. A. Durgin, Assist-
ant Chief Testing Engineer of the Com-
monwealth Edison Company, gave the
first of a series of 20 minute talks on
the "Fundamentals of Illumination."
Sixty-four members and guests attended
the meeting.
The December meeting is to be held
at the residence of Mr. W. A. D. Curtis,
when a paper entitled "The Lighting of
the Home" will be discussed.
In January, there will be a joint meet-
ing with the Chicago Architectural Busi-
ness Men's Association, at which Mr.
J. B. Jackson, secretary of the Chicago
Section, will give a paper entitled
"Planning Lighting Installations."
NEW ENGLAND SECTION
A meeting of the Board of Managers
of the New England Section was held
in the Hotel Georgian, November 3.
Those present were : C. A. B. Halvor-
son, chairman; C. M. Cole, secretary;
J. \Y. Cowles, vice-president; R. B.
Hussey, J. M. Riley and R. C. Ware.
The meeting was devoted to the con-
sideration of a program of meetings for
the present year. A campaign is to be
undertaken to increase the membership
of the section.
NEW YORK SECTION
A meeting of the Board of Managers
was held November 15, in the general
office of the society, 29 West 39th Street,
New York City. Those present were:
W. C. Morris, chairman; G. H. Stick-
ney, vice-president; O. H. Fogg, W. H.
Spencer, H. B. McLean, M. D.
McDonald, H. B. Rogers, S. W. Van
Rensselaer, and C. L. Law, secretary.
The meetings' program for the rest of
the year was discussed. Arrangements
have been made for a number of excel-
lent papers and joint meetings with
other societies.
The New York Section held a joint
meeting with the New York Companies'
Section of the National Electric Light
Association in the Auditorium of the
New York Edison Company, November
17. Mr. S. G. Rhodes of the New York
Edison Company gave a talk on "Street
Lighting Abroad." Mr. H. W. Jackson
of the General Electric Company gave
a lecture on the latest improvements in
incandescent lamps. Mr. Alexander
Maxwell of the New York Edison Com-
pany exhibited several Neon tube lamps.
About 125 members of both organiza-
tions were present.
TRANSACTIONS I. E. S. — PART I
PHILADELPHIA SECTION
On November 20, a joint meeting of
the Philadelphia Section was held with
an ophthalmological society of Philadel-
phia.
The following is the program for
future meetings :
Monday, December 8.
Joint Meeting with Philadelphia Section
A. I. E. E.
"Brightness Measurements versus Illu-
mination Measurements."
By Dr. Herbert E. Ives.
"Railway Car Lighting."
By Air. Geo. H. Hulse.
"The Mercury Quartz Tube Lamp."
By Mr. Buckman.
Friday, January 6.
"Deficiencies of the Method of Flicker
for the Photometry of Lights of
Different Colors."
By Prof. C. E. Ferree.
Saturday, February 7.
Meeting under the Auspices of Drexel
Institute.
"Light and How to Use It."
By Mr. C. O. Bond, President
of I. E. S.
Wednesday, February 18.
Joint Meeting with Franklin Institute.
"Artificial Daylight."
By Dr. Herbert E. Ives.
Friday, March 20.
"Lighting and Signalling Systems of
Subways."
By Mr. F. D. Bartlett.
"The Sun— The Master Lamp."
By Prof. James Barnes.
Thursday, April 9.
Joint Meeting with Franklin Institute.
"Recent Developments in the Art of
Illumination."
By Air. Preston S. . Millar.
Friday, April 17.
"The Structure of the Normal Eye and
its Ability to Protect Itself Against
Ordinary Light."
By Dr. Wendell Reber.
"Glassware for Illumination and Other
Purposes."
By Mr. James Gillinder.
Friday, AIay 15.
Mass Aleeting of all the Engineering
Societies of Philadelphia and
Vicinity.
Special Program to be arranged and to
include an address on
"The Relation of Engineers to the
Progress of Civilization."
By Dr. Chas. Proteus Steinmetz.
PITTSBURGH SECTION
A meeting of the Pittsburgh Section
was held October 17 in the Auditorium
of the Engineering Society of Western
Pennsylvania. Sixteen members and
guests were present. Air. L. L. Hopkins
reviewed the proceedings of the Pitts-
burgh Convention and Dr. H. H. Turner
gave a paper entitled "The Essential
Elements of Vision." Dr. Turner's
paper was supplemented with a series
of slides and models.
The following program of meetings
has been announced tentatively:
November — "Technical Discussion of
the Elements of Lighting" by Prof.
Hower and others.
December — A joint meeting with the
Pittsburgh Section of the Ameritan
Institute of Electrical Engineers. A
Central Station paper will be pre-
sented by H. N. Muller of the
Duquesne Light Company.
January — A paper to be selected by the
members from Cleveland. The sub-
ject will be announced later.
February — "Railroad Yard Lighting" by
A. C. Cotton and A. Kirschberg of
the Pennsylvania Railroad Com-
pany.
TRANSACTIONS I. K. S.— PART I
March — A gas lighting subject; the
speaker to be announced later.
April — "Developments of Flame Carbon
Arc Lamps" by C. E. Stephens.
May — "Store Lighting" ; speaker to be
announced later.
June — Open.
New Members.
The following twelve applicants were
elected members of the society at a
meeting of the Council, November 14,
I9I3-
De Vine. H. C.
Manager, Pittsburgh Lamp, Brass &
Glass Company, 731 Arch Street,
Philadelphia, Pa.
Fitch, W. S.
Construction Engineer, Dennison
Mfg. Company, Framingham, Mass.
Hageman, Jacques R. G.
Engineering Department, Bell Tele-
phone Company, 2129 Ritner Street,
Philadelphia, Pa.
Hass, Henry P.
Chief Inspector, Department of
Tests, N. Y. N. H. & H. R. R. Com-
pany, New Haven, Conn.
Hicks, Leslie R.
Superintendent, Fall River Electric
Light Company, 14 Bedford Street,
Fall River, Mass.
Hostetter, John S.
Manager Fixture Department, Bar-
den Electric & Machinery Company,
in Main Street, Houston, Tex.
Johnston, R. J.
Testing Department, General Elec-
tric Company, Schenectady, N. Y.
Jordan, Horace W.
Illuminating Engineer, Edison Elec-
tric Illuminating Company of Bos-
ton, 39 Boylston Street, Boston, Mass.
Mercer, J. M.
Car Lighting Engineer, The Adams
& Westlake Company, 319 West
Ontario Street, Chicago, 111.
Osborn, Frederick A.
Professor of Physics, University of
Washington, Seattle, Wash.
Staley, V. L.
General Electric Company, Harri-
son, N. J.
Taylor, Frank C.
Assistant in Electrical Engineering
Department, Rochester Railway &
Light Company, Rochester, N. Y.
Additional Sustaining Members.
The following organizations were re-
cently elected sustaining members:
Gill Brothers Company.
101 Park Avenue, New York, N. Y.
Official Representative: John Beis-
wanger.
Jefferson Glass Compa: y.
Follansbee, W. Va.
KOERTING & MATHEISEN.
22 East 21st Street, New York, N. Y.
Official Representative: Charles
Arnold Chapin.
The Electric Power Company, Ltd.
506 Confederation Life Building,
Toronto, Can.
Official Representative : Wills Mach-
bachlan.
Personal.
Mr. C. O. Bond, president of the
Illuminating Engineering Society, was
recently awarded the Beal medal of the
American Gas Institute for the best
paper read at the 1912 Convention of
that organization. Mr. Bond's paper,
entitled "Photometry of Incandescent
Gas Lamps," appears in the 1913 Pro-
ceedings of the Institute.
TRANSACTIONS I. E. S. — PART I
Mr. W. H. Gartley, a past president
of the Illuminating Engineering Society,
was recently elected President of the
American Gas Institute.
Hollis Godfrey, Ph. B., Sc. D., F. R.
G. S., was recently elected president of
the Drexel Institute of Art, Science and
Industry, Philadelphia, by the trustees
of that institution. He will assume the
presidency December I, 1913. Dr. God-
frey is well known as an educator, a
business man and engineer. He organ-
ized the department of science in the
High School of Practical Arts in Bos-
ton, and for four years served as its
head. He also spent six years in night
school work in Boston along the same
lines which the Drexel Institute is being
conducted. With two other leaders in
education he organized the Garfield
school, and for two years directed its
policy in the teaching of science, and
in extension work. For three years he
served as a member of the Board of
Visitors of Tufts College, and was a
member of the alumni council of the
Massachusetts Institute of Technology
for two years. He has outlined a course
in industrial engineering for the Society
of Promotion of Engineering Education,
and has been a lecturer and consultant
in a number of educational institutions,
among which are Dartmouth College,
Simmons College and the University of
Wisconsin. For several years past he
has been chief of the Bureau of Gas of
Philadelphia, and recently devoted an
exhaustive study to the organization and
operation of the water bureau of the
latter city. In addition to outlining a
new lighting plan for Atlantic City, he
has also been a consultant in the restora-
tion of the lighting of Independence
Square, Philadelphia. He is the author
of a book on sanitary engineering en-
titled "The Health of the City," two
books on "Chemistry," and a number
of monographs on scientific subjects.
Besides having membership in a num-
ber of clubs Dr. Godfrey is a member
of the Phi Beta Kappa and the Theta
Delta Chi fraternities, the American
Society of Mechanical Engineers, the
American Public Health Society and
the Illuminating Engineering Society.
Obituary.
George H. Hoffman, a district man-
ager of the Philadelphia Electric Com-
pany, died in Philadelphia, November 3,
after a brief illness. He was born in
New York City, December 22, 1848, and
received a public school education.
From 1873 to 1877 Mr. Hoffman was a
member of the New York School
Board, and from 1882 to 1889 was a
member of the City Council. He was
appointed by President Cleveland As-
sistant United States Appraiser in 1885,
and served in this position until 1889.
Later he engaged in the wool business
and was also connected with the Nord-
Amerika, a German newspaper. In
June, 1895, he identified himself with
the electric lighting industry as mana-
ger of the West End Electric Light
Company of Philadelphia. In 1901,
when the West End, Columbia,
Diamond and Wissahickon companies
were consolidated into the Philadelphia
Electric Company, he was made man-
ager of the northwestern district of the
company. He was a charter member
of the Pen and Pencil Club of Phila-
delphia, and had served as vice-president
of the International League of Press
Clubs. He took an active interest in the
affairs of the Philadelphia Section of
the Illuminating Engineering Society
besides being a member of several
other societies.
TRANSACTIONS
OF THE
Illuminating
Engineering Society
NOVEMBER, 1913
PART II
Papers, Discussions and Reports
[ NOVEMBER, 1913 ]
CONTENTS - PART II
The Cooling Effect of Leading-in Wires upon trie Filaments
of Tungsten Incandescent Lamps of the Street Series
Type. By T. H. Amrine 385
Some Theoretical Considerations of Light Production. By
W. A. Darrah 400
The Pentane Lamp as a Working Standard By E. C.
Crittenden and A. H. Taylor 410
Experiments in the Illumination of a Sunday-School Room
with Gas. By Edwin F. Kingsbury 439
Characteristics of Enclosing Glassware. By Van Rensselaer
Lansingh 447
The Photo-Electric Cell in Photometry. By F. K. Richt-
myer 459
Factory Lighting. By M. H. Flexner and A. O. Dicker • . 470
Hospital Lighting. By William S. Kilmer 488
Store Lighting. By J. E. Philbrick 499
Distinctive Store Lighting. By Clarence L. Law and A. L.
Powell 515
Recent Improvements in Incandescent Lamp Manufacture.
By Ward Harrison and Evan J. Edwards 533
V4*
/
THE COOLING EFFECT OF LEADING-IN WIRES UPON
THE FILAMENTS OF TUNGSTEN INCANDESCENT
LAMPS OF THE STREET SERIES TYPE.*
BY T. H. AMRINE.
Synopsis: This study of the cooling effect of the leads upon the
filaments of street series lamps was undertaken as a part of a general
investigation into the effect that dimensions and material of lead wires
and supports have upon incandescent lamp design. The method used was
to calculate the average per cent, of normal candle-power and average
per cent, of normal wattage over the cooled portion of the filament by
means of measurements made upon two sets of lamps of exactly the same
construction except having different filament lengths. The variation of
the cooling effect with lead material, lead diameter, lead length, filament
diameter and filament material was determined. The cooling effect of
any lead upon any filament was shown to be dependent mainly upon
(i) the resistance to heat flow presented by the leads and the cooled por-
tion of the filament, (2) the diameter of the filament and (3) the maxi-
mum temperature of the filament, i. e., the temperature of the uncooled
portion. For lamps having lead and filament dimensions encountered in
street series lamps the cooling effect decreases with increase of lead
length, with decrease in lead diameter and with increase in thermal
resistance of the material of the lead. With lamps having the same lead
construction but with filaments of different diameters the cooling effect
shows a maximum at a value of filament diameter which is dependent
upon the lead construction used.
The accuracy with which it is possible to predetermine the
dimensions of the filament of a lamp having the desired candle-
power, wattage and construction is at present very largely
limited by the lack of knowledge of the amount of cooling
effect of lead wires and supporting anchors. A length of drawn
tungsten wire of a certain diameter and carrying a certain cur-
rent will, when subject to no cooling effect, consume a definite
number of watts per centimeter length and will produce an
equally exact, if not so readily determined, candle-power per unit
length. The determination of the total wattage and total flux
* A paper read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
386 TRANSACTIONS I. E. S. — PART II
of light from such a filament of any length would, therefore,
involve only multiplication and a correction for the light absorbed
by the bulb. The presence of leading-in wires and anchors sub-
jects the filament to a cooling effect which is dependent upon
their material and dimensions and upon the number of anchors.
This cooling effect makes the average watts per centimeter and
candle-power per centimeter of a given filament dependent not
solely upon the dimensions of the filament and the current flow-
ing through it, but also upon the construction of the lamp as
regards leads and anchors.
The work reported in this paper was undertaken as a part
of a general investigation into the effect that dimensions and
material of lead wires and supports have upon incandescent
lamp design. The study of the cooling effect of the leads in high
current, low-voltage lamps, such as street series lamps was taken
up first on account of the fact that the leads of these lamps cause
heat conduction losses that are large and relatively easily meas-
ured. An investigation into the effect which changes in such
variables as lead diameter, lead length, material of lead, filament
diameter, filament length and filament temperature, can, there-
fore, be carried on with such lamps and data obtained which is of
value not only in connection with the design of lamps of this
type, but also in that it will furnish some information as to the
general laws governing the relation between cooling effect and
the variables mentioned. This information as to the approxi-
mate nature of these laws will serve as a guide in the investiga-
tions upon the ordinary multiple lamp where the cooling effect is
less marked and less easily measured.
Hyde, Cady and Worthing have published* the results of an
investigation into the energy losses in lamps of the multiple type
and have shown the variation in the conduction losses due to
change in filament material and filament temperature. The
variations due to lead dimensions and material and filament di-
ameters were not determined in this investigation, and no data
seems to have been published on this phase of the subject.
* A Study of The Energy Losses in Electric Incandescent Lamps. E. P. Hyde, F. E-
Cady, A. G. Worthing. Trans. I. E. S., Vol. VI., No. 4, page 238.
AMRINE: C00UNG EFFECT OF LEADING-IN WIRES 387
METHODS OF INVESTIGATION.
A number of methods of carrying out this investigation are
possible and were considered before the work was taken up.
A strictly mathematical method can, of course, be employed, the
temperature of the filament at different distances from the
leads being calculated and the total changes in candle-power and
wattage due to the cooling effect determined therefrom. On
account of the number of variables and complexity of the prob-
lem, the mathematical method yields a rather unwieldy equation.
Its accuracy is dependent upon the thermal constants of tungsten
at high temperatures and upon the law of their change with the
temperature and upon the thermal resistance of the welded
joints between the lead wire and filament and other factors upon
which there is no good data. A much preferable method is to
measure directly the temperature of the filament at points at
different distances from the leads by some means such as a
thermocouple made of very fine wires or by an optical pyrometer
such as was used by Hyde, Cady and Worthing. These tempera-
ture measurements are laborious to make by either method and
the time required to cover the ground which was desired to cover
in this investigation would be very great. By the optical pyrom-
eter method measurements of temperature over the first two
or three millimeters of the filament adjacent to the lead of a
lamp of the series type are very unsatisfactory on account of the
very low luminosity of the filament at these points. For these
reasons the methods involving temperature measurements were
not adopted, though for the purposes of studying the temperature
conditions adjacent to the leads a considerable number of tests
were made by a method only slightly modified from that used
by Hyde, Cady and Worthing. The modification was simply
the interchanging the positions of the lamp under test and the
comparison lamp so as to permit of the latter burning at a lower
temperature.
The method adopted was to arrive at the figure for the
average per cent, of normal wattage and average per cent, of
normal candle-power over the cooled portion of the filament
by means of measurements made upon two sets of lamps having
different lengths of filaments. This method was made possible
388 TRANSACTIONS I. E. S. — PART II
by the fact that modern tungsten wire drawing methods permit
of the manufacture of wire of almost exactly uniform diameter
and composition throughout its length. The average candle-
power and wattage of two sets of lamps of exactly the same con-
struction can, therefore, be depended upon to be very nearly the
same. Also the difference between the average candle-power,
for instance, of a set of lamps of a certain construction and that
of another set of exactly the same construction, except for fila-
ment length, will be the candle-power of an uncooled portion of
the filament equal in length to the difference between the average
filament lengths of the two lots of lamps.
Assume that it is desired to determine the cooling effect upon
a filament of a given diameter of a lead of given dimensions
and material. Two lamps are made from wire of the correct
diameter taken from the same spool and having the desired
dimensions and material of leads and of exactly the same con-
struction throughout, except that one has a filament of length 1
and the other a filament of length 1'. Let c and w represent
the candle-power per centimeter length and the watts per centi-
meter length respectively of the uncooled portion of the filament
with a current passing through the lamp which will bring this
uncooled portion of the filament to the desired temperature —
say that corresponding to normal operation. Let a equal the
total length of the cooled portion of the filament in each case.
This will be the same for two lamps of similar construction ex-
cept for filament length, if the filament is longer than a certain
minimum length. The expressions for the total candle-power
and total watts of the two lamps will be
C=aK-|-(l — a)c, for lamp of filament length 1
W=aP-f-(l — a)w, for lamp of filament length 1
Similarly,
C'^aK-l-O' — a)c, for lamp of filament length 1'
W'=aP+0' — a)w, for lamp of filament length 1'
where K is the average per cent, candle-power and P the average
per cent, wattage over a cooled portion of the filament a centi-
meters long. These figures for average per cent, are based upon
the candle-power and wattage of a centimeters of the uncooled
portion of the filament.
AMRINE : COOLING EFFECT OF LEADING-IN WIRES 389
C and C can be determined by careful photometer measure-
ments and W and W by means of a potentiometer. An investi-
gation with an optical pyrometer showed that in none of the
lamp constructions which it was planned to investigate was there
any cooling effect at a distance greater than two centimeters. For
the sake of uniformity this distance of 2 centimeters was adopted
arbitrarily as the length of the cooled portion adjacent to
one lead. Hence a is equal to 4 centimeters. Between the above
equations, K and P and the values of c and w can be determined,
since 1, 1', C, C, W, W and a are known. The values of K and
P based on the standard distances of 2 centimeters from each
lead, are measures of the cooling effect. By using these values
and the values of c and w, as determined above, one can calculate
the candle-power and watts of an ideal filament {i.e., one which
is not subject to a cooling effect) of length 1 or 1' and compare
the values obtained with C, C, and W and W.
A quantity M was also calculated for each case studied. This
quantity is the per cent, difference between the wattage of an
ideal lamp with an uncooled filament of 10 centimeters length
and that of a lamp of any given lead construction of the same
candle-power and having a filament of the same diameter and
operating at the same current. M expresses by one quantity the
result of cooling effect on both the candle-power and watts of a
filament.
This method has the advantage that the work required for
one determination can be carried out in a reasonable length of
time so that the various experiments can be made upon a suffi-
cient number of lamps so that a good average result can be ob-
tained. In this work on an average of four lamps of each con-
struction were investigated so that it is felt that fairly good
average values were obtained.
EXPERIMENTAL RESULTS.
The lamps used in the experiments were similar in construction
to the regular street series tungsten lamps except that there were
no anchors or supports used. All lamps had drawn wire tungsten
filaments and were made in straight sided bulbs of 4^ inches
(11. 11 centimeters) maximum diameter (commercially known
39° TRANSACTIONS I. E. S. — PART II
as S-35 bulb) and with a skirted screw base (commercially known
as No. 108 base) and with the same size of glass stems. The
dimensions of the lead wires, of course, were different in the
different experiments. Fig. 1 shows the type of lamps experi-
mented upon.
The routine followed in carrying out these experiments was as
follows : Assuming, for example, it was desired to compare the
cooling effect on a filament of a given diameter and tempera-
ture, of leads of two different diameters. Ten lamps were made
up with filaments taken from the same sample of wire with
each of the desired lead constructions, all lamps of each lot
of ten being exactly alike except that five had filaments of about
6 centimeters length and five had filaments of about 15.5 centi-
meters length. After the filaments were mounted they were
each carefullly measured before sealing-in for filament length
between leads. The lamps were exhausted in the regular manner,
based and then all carefully measured, at the amperes to give
the desired temperature, for candle-power upon a precision
photometer and for wattage by means of a potentiometer. The
values for K, the average per cent, of normal candle-power,
and P, the average per cent, of normal wattage each over a
portion of filament to a distance of 2 centimeters from each lead,
were then calculated as shown above. From these values the
value of M, the per cent, wattage change due to cooling, were
calculated. The average figures obtained with each lot of lamps
were used as the correct values of K, P and M.
On account of the fact that the accuracy of the wattage meas-
urements is considerably better than that of the candle-power
measurements, the curves shown were based upon the wattage
measurements. From the relation between K and P determined
from all the experimental data, the points on the curves for K
as plotted were calculated from the values of P. In this way
the K curves were "smoothed out" and made to conform with the
P curves. By this plan it was thought that the relation between
K and different variables as lead length, lead diameter, etc., are
more accurately shown and that the absolute values of K, as
plotted, are probably as accurate as the individual experimental
values obtained.
3qO
^
**■
t.
yv
T jl^
I
Fig. i.
M K
1 1.0 62
56
10.0 54
52
9.5 50
48
9.0 46
\M
^
,j
"" "s.
K .
^^^
■'*
T ~
CURVES SHOWING COOLING EFFECT OF
LEADSOF DIFFERENT LENGTHS UPON
A FILAMENT .01065 DIAM. CARRYING fc5
AMPjS. LEADS OF COPPER .05;DIAM.
LENGTH IN INCHES
62
10.5 60
54
a5 52
^\p
p
>~^
^^_
K
~~.
--.^
^><C
Hr+N
M,
*■»«!
^
s^\
CURVES SHOWING COOLING EFFECT OF LEADS Oc "^^
DIFFERENT D ivE-F-=s .":'. A r _A".'E\T.OIQ6r> \
DIAMETER CARRYING 6.5 AVP LEAD: C^COP-E3 := _:'.;. >
>
81
LEAD DIAMETER
Fig. 3-
A
r
^\
1/
/K
frv
:jr.
0F.0
ES SHOWING COOLIN
y«8 CU LEADS UPON Fl
jEFF
-AME
-A
i-s\
X
•
^».
— s?
,^
.003 .010 .Oil .012 .013
FILAMENT DIAMETER IN INCHE5
Fig. 4.
AMRINE : COOLING EFFECT OF LEADING-IN WIRES 391
Fig. 2 shows the values obtained at operating efficiency for
K, P and M for lamps having different lengths of leads. In
these and in the following curves K is the average per cent,
candle-power and P the average per cent, wattage over the two
centimeters of the filament adjacent to each lead; that is, the
section of the filament which embraces the cooled portion. These
values are based upon the candle-power and wattage of two cen-
timeters of a similar but uncooled filament carrying the same
current. M is the per cent, difference in wattage between an
ideal filament and a cooled filament as defined under the heading
of Method of Investigation. These curves indicate that for the
lead construction used the longer the lead, the less the cooling
effect as is shown by the higher values of K and P and the lower
values of M. The lead length as used in this connection is taken
as the length of lead from filament to the glass stem.
Fig. 3 shows the greater cooling effect of large diameter leads
as compared with small diameter leads of the same material.
In order to compare the relative cooling effects of copper and
nickel leads of the same dimensions, K, P and M were deter-
mined for a 6.5 ampere filament using leads 0.050 in. in diameter
and iy2 in. (3.81 cm.) long in each case. The values obtained
were as follows :
p K M
Copper 82.6 56.3 10.52
Nickle 90.6 67.5 9.24
The influence of filament diameter upon cooling effect is shown
by the curves of Fig. 4. The rather unexpected shape which
these curves take is no doubt accounted for by the fact that the
total resistance to heat flow from the filament is made up of
three parts, that due to the leads, that due to the welded joints and
that due to the cooled portion of the filament. With the same
size leads and the same sort of a welded joint this total resistance
might be expected to show a minimum at some value of filament
diameter. In a lamp of the construction studied this minimum
occurs at about 0.0108 in. diameter.
Fig. 5 is a set of curves which shows the variation in cooling
effect with the filament temperature. These curves are plotted
between "per cent, of normal -= — -" and "percent, of normal P,"
^3/2
392
TRANSACTIONS I. E. S. — PART II
"per cent, of normal K and "per cent, of normal M." The
quantity ^— r is a measure of filament temperature, I represent-
ing the current through the filament and d the diameter of the
filament. A value of -=— p equal to 0.0059 was taken as normal.
This corresponds to approximately 1 watt per projected candle-
110
M
1
95
P.
V
90
K/
CURVES SHOWING VARIATION Op % OF NORMAL P.K &M Vl
ViS BASED ON A VALUE OF Virt CORRESPONDING TO i.oc
WATT/PROJECTED CP AS NORMAL.
ITU
| |
'/d* * 1000
Fig. 5-
220O
2100
2000
) 1900
c
; 1800
\ 1700
2 isoo
J
51500
J
i|400
1300
■
|
k.
3.2 AMR
7.5 AMR
1
V
6.5 AMP
5.5 A
■;p
CURVES SHOWING- TEMPERAIUKtb IN
DES.C. OF A FILAMENT AT VARIOUS AMPS.
AND VARYING DISTANGE5FR0M THE LEADS._
7.5A^
RNOf
MAL
06V l|
OJ.L
""""O 2 4 6 8 10 12 W 16 18 20 22 24 26
DISTANCES FROM LEAD IN MM.
Fig. 6.
power. It also corresponds approximately to the filament tem-
peratures used in the measurements made upon the relative cool-
ing effects of leads of different diameters, lengths and material.
These curves indicate that with higher filament temperatures,
i. e., lower watts per candle-power, the cooling effect is less
AMRINF : COOLING EFFECT OF LFADING-IN WIRES
393
marked. This can be readily noticed in a lamp if the cooled por-
tion of a filament is observed first with a large current flowing
and then with a smaller one. In the latter case the filament is
cooled out to a considerably greater distance from the lead. The
curves of Fig. 6, the data of which was obtained by the optical
pyrometer method, show the decrease in temperature near the
leads and the greater cooling effect at higher temperatures.
A study of the foregoing data shows that the cooling effect of
any lead upon any filament is dependent mainly upon three things :
(i) the resistance to heat flow presented by the leads and the
cooled portion of the filament, (2) the diameter of the filament
and (3) the maximum temperature of the filament, i. e., the
Kooo
12000
10.000
fcooo
CURVES SHOWING RELATION BETV
TOTAL THERMAL RESISTANCE (LT.J36F
L" THERMAL RESISTANCE OF LEAD.
'EEN_
AHD|.
TF- - " OFT" FILAM
RELATIVE THERMAL C0NDUCTIV1T1E
1 CU.-5.07. NICKEL-.83G. TUNCSTEN-2
:NT.
8-
2i
O
<>
tOO 500
I. +.256 F
Fig. 7-
temperature of the uncooled portion. With a given maximum
filament temperature and a given thermal resistance due to the
leads and the cooled portion of the filament the same number of
watts will be lost by conduction (neglecting differences due to
the amount of heat radiated by the lead and cooled portion of
the filament). With a given number of watts escaping by con-
duction there will, of course, be a greater per cent, reduction in
wattage with a filament of small diameter than one with a fila-
ment of a larger diameter. By plotting the data obtained with
"total thermal resistance," that is, the resistance of the lead plus
that due to the cooled portion of the filament, as abscissae and
— as ordinates, where d is filament diameter, an approximately
394
TRANSACTIONS I. E. S. — PART II
smooth curve is obtained. In this curve, Fig. 7, the "total thermal
resistance" is taken to include 0.236 in. of the filament and is
calculated using lengths in inches, diameters in inches and the
relative thermal conductivity of the materials as follows :
Copper 5.07
Tungsten 2.26
Nickel 0.876
The curve includes the data taken upon leads of different
diameters, lengths and materials and different filament diameters
and should show, over the range covered by the experiments,
approximately the cooling effect of any lead upon a filament of
any diameter. It does not, however, take into consideration the
influence of change in thermal conductivity with temperature,
^88
!-
z
<
L
f
$&-
40
4
4
f
,f
9
i*
j= 30
r 88
CURVE
20C-E-
SSHO
SERIES
viks'h
JMPH
EATINf
TH.O?
U/OF40AMR
ijcu LEADS.
Fig. 8.
effect of radiation from the leads, etc. The filament temperatures
correspond to -=— 7- equal to 0.0059, tnat is. approximately one
watt per projected candle-power. By the use of this curve and
the P curve of Fig. 5, one can determine approximately the
values of P for filaments at other temperatures. Knowing that
K varies about as the cube of P the value of the average per cent,
candle-power over the cooled portion of a filament with any lead
construction can be determined.
The data obtained in the experiments described show that there
is a marked cooling effect -in lamps of the construction used. That
this cooling effect is of vastly more than mere scientific interest
in the case of these lamps is, perhaps, not generally recognized
AMRINE : COOLING EFFECT OF LEADING-IN WIRES 395
except by lamp specialists. However, in the design and rating of
and measurements upon these lamps this cooling effect must be
taken into practical consideration.
In Fig. 8 are shown curves which demonstrate that the cooling
effect of leads must be taken into consideration in all careful
measurements of series lamps. These curves were obtained by
holding the normal current through the lamp while the voltage
and candle-power were measured at short intervals of time. Due
to the cooling effect of the leads, the lamp does not come up to
full normal candle-power until several minutes after the current
is turned on. Errors that would be quite serious in careful work
are caused if series lamps are photometered or measured for
volts at amperes without taking the precaution first to allow the
:'5
<KL7
1
1 1 1 1 1!
CURVES SHOWING AVE?i3F.
EF* C_= V .» r.AMEVS CF WFTTim
15 AJ
■=:.
UEM
.:-:'■
V
Fig. 9.
lamp to heat up thoroughly. For instance, an error of about
1.3 candle-power or 6.5 per cent, would be caused in photometer-
ing a 20 candle-power, 4.0 ampere series lamp 20 seconds after
lighting up instead of allowing it to heat up thoroughly.
Due mainly to this influence of the cooling effect of leads, it
is impracticable to make, for instance, 7.5 street series lamps of
the various candle-power sizes at the same efficiency rating.
Lamps of the same current rating have, of course, a candle-
power almost proportional to the filament length. The loss in
candle-power and the decrease in wattage for a low candle-power
lamp of a given diameter of filament and a given current flowing
is a much larger per cent, of the total for a short filament or low
candle-power lamp than with a long filament or high candle-
39^
TRANSACTIONS I. E. S. — PART II
power lamp. Hence, if the current in the short filament lamp is
increased until the filament is operating at the same watts per
mean spherical candle-power then its temperature out on the
uncooled portion of the filament is much greater than the maxi-
mum temperature of the long filament lamp and will, therefore,
give a much shorter life. In Fig. 9 is a curve which shows the
watts per mean spherical candle-power of lamps made with fila-
ments of the same diameter and with the same current passing
through them, but of different filament lengths. It shows that
at the same amperes a short filament is operating at a very much
poorer efficiency than a long filament and if operated at the same
efficiency would give a very poor life.
&o
J07.8
<ia
\ = .-::- .<,-•!*: a^e^es an; ~ ;_-- . E '.. .:
\ |0F SERIES LAMPS HAVIN6 FILAMENTS OF DIFFERENT
HORIZONTAL CANDLE-PCWE8. |
FILAMENT .Olia'DIAM.- .OS'* I5CU. LEADS
/r
<£
\
_)
^V/
|
zoo
' 0 Z 1 <c 8 10 12 H IS IS 20 22 24 26 28 30
CM. LENGTHS
Fig. io.
In Fig. io is shown what would be the effect if street series
lamps of a given wire size and current with their great cooling
effects were rated at the same watts per mean horizontal candle-
power regardless of their filament lengths. These curves assume
7.5 ampere lamps of different candle-powers, that is, different
filament lengths. They are assumed all to be rated at i.i watts
per mean horizontal candle-power. The curve marked "amperes"
shows the amperes that it would be necessary to pass through
the filaments of the various lengths in order to bring them to the
same mean horizontal efficiency. The great increase in amperes
necessary to bring the short filaments up to efficiency, of course,
very seriously cuts down their lives, since a lamp's life is deter-
mined by the temperature of the hottest part of the filament.
This decrease in life is shown by the curve marked "life." The
AMRINE: COOLING EFFECT OF LEADING-IN WIRES
397
character of these curves is almost wholly brought about by the
cooling effect of the leads, though the change in spherical reduc-
tion factor with the filament length enters to a minor extent.
By going to the extremes in cooling effect by using filaments
so short that they are cooled over practically their entire length,
one can arrive at the unusual condition of having two lamps with
filaments of the same diameter and differing in length by 30
per cent, or more but with the same voltage and candle-power.
This is brought about by the difference in the volt candle-power
characteristic caused by the difference in the cooling effect of the
leads on the two filaments. Fig. 11 shows the volt candle-power
A,
/
•
/
f y
/a
£
/
/ y
1
4/
vy
Z
22
^
*
:--
VOLT CANDLE-POWER CHARACTERISI
6£AMP LAMPS wrrH.05V^CU. LEADS
cor
B-
1.59
VOLTS
Fig. 11.
characteristics of two lamps made exactly alike except for fila-
ment length. Lamp A had a filament 1.31 cm. long, lamp B,
1.59 cm. long. Both filaments were made of wire 0.0118 in. diam-
eter taken from the same spool. The characteristics of the two
lamps intersect and show that both lamps give 2.27 candle-power
at 1.44 volts. In other words, if one had a lamp of the construc-
tion of lamp B, he could remove 17.5 per cent, of the filament
and still have a lamp of the same candle-power and voltage.
Notwithstanding the magnitude of the cooling effect on lamps
of the street series type it is for the most part impracticable to
let this factor determine the lead construction of these lamps.
That is, it is usually not feasible in practice to so construct a
lamp that the losses due to cooling effect approach a minimum.
Other conditions will usually limit the designer's choice of lead
39§ TRANSACTIONS I. E. S. — PART II
dimensions and materials. The appearance of the lamp, bulb
dimensions and added energy losses due to electrical resistance
will usually limit the length of leads. The vapor pressure, and
electrical resistance of the material of the lead will usually deter-
mine its diameter and these factors as well as the ease with which
the metal may be subjected to the various manufacturing pro-
cesses will determine the material of the lead. A knowledge of
the manner in which the cooling effect varies with lead material
and dimensions is therefore chiefly of importance in that it
enables one to make the proper allowances in candle-power, watts
and volts in designing a lamp to a given specification.
The author wishes to acknowledge his indebtedness to Mr.
L,. M. Moss, who assisted in all the experimental work and in
the preparation of the data for this paper.
DISCUSSION.
Mr. Evan J. Edwards : I have consulted Mr. Amrine regard-
ing a demonstration which was used in the presentation of the
paper "Recent Improvement in Incandescent Lamp Manufac-
ture," and I believe it worth while to repeat it, for it shows in a
very striking manner the cooling of the filament adjacent to sup-
ports and leading-in wires.*
There is one interesting point concerning the cause of the cool-
ing which Mr. Amrine has not emphasized, and that is the
changed rate at which electrical energy appears as heat in the
cooled portions. The fundamental cause of the reduced tem-
perature is, of course, the conduction of heat to the supports or
leading-in wires. But the reduced temperature condition, in the
case of metal filaments, brings about a lowering of the resistance
which also contributed to lowering of the temperature. The cur-
rent in all portions of the filament is the same, and therefore,
the rate at which heat is received by any portion of the filament
is proportional to the resistance of that portion.
Changed resistance therefore has the effect of helping to lower
the temperature of the cooled portion of metal filaments because
* A stereopticon was used to project the direct image of the lighted filament of a
large special lamp on a background lighted by the regular light source of the lantern.
After adjustments of current in the special lamp were made, the hottest parts of the fila-
ment appeared as bright lines and the cooled portions as dark lines.
COOLING EFFECT OF LEADING-IN WIRES 399
of their positive resistance-temperature coefficient. The oppo-
site is true for the negative coefficient carbon filaments.
Mr. M. Luckiesh: At the bottom of the tenth page Mr.
Amrine states "These curves indicate that with higher filament
temperatures, i. e., lower watts per candle-power, the cooling
effect is less marked." This is at once evident from the funda-
mental laws of radiation. As the temperature of the filament
increases, the conduction loss becomes proportionately less
because it increases approximately directly with the temperature
while the candle-power is a function of a higher power of the
temperature. Therefore as the temperature of the filament is
increased the candle-power will increase much more rapidly than
the conduction losses in the leading-in wires.
Mr. T. H. Amrine: I wish to add that the lamp with the
shorter filament, lamp A, has a wattage of 11.07 and lamP B has
a wattage of 12.25 at the voltage which makes the candle-powers
of the two lamps equal, that is, at 1.44 volts.
400 TRANSACTIONS I. E. S. — PART II
SOME THEORETICAL CONSIDERATIONS OF LIGHT
PRODUCTION.*
BY W. A. DARRAH.
Synopsis: This paper sets forth some of the inherent limitations of
the various electric illuminants now in use, and discusses briefly the effect
of these limitations upon the progress of the art of illumination. The
basic theory of light production by the acceleration of electric charges
is analysed with a view to indicating the relative possibilities of the
various types of illuminants particularly the arc and the incandescent
lamp. The structure of the atom is considered in the light of modern
theories, and some general deductions made regarding the properties
which a substance is likely to exhibit if well adapted for use as a radiating
body. A tentative explanation of the action of selective radiation is
given, and of the part which this phenomenon plays in present illuminants.
It is demonstrated that to secure high luminous efficiencies selective
radiation must be relied upon, and therefore unless new materials are
discovered which exhibit selective radiation, while in the solid state, the
efficiency of the electric arc will remain materially higher than the efficiency
of the so-called incandescent lamp, where electrical energy is trans-
formed into heat and then into light by means of the resistance of a
conductor.
It is proposed in this paper to discuss some theoretical phases
of light production with a view of indicating how these consid-
erations have influenced the progress of the art of illumination
and their probable effect upon future developments in the direc-
tion of higher efficiencies.
No claim is made here for original theories, as modern con-
ceptions and methods of mathematical analysis have been used
freely; but an effort has been made to apply these theories in a
way to indicate the trend of engineering progress.
The generally accepted, and most satisfactory conception of
light, is. that of a transverse electromagnetic vibration, of the
ether, which travels in straight lines at a speed of approximately
180,000 miles per second. Having identified light as an elec-
tromagnetic vibration it is necessary to differentiate it from other
* A paper read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
DARRAH : SOME CONSIDERATIONS OF EIGHT PRODUCTION 4OI
electromagnetic vibrations, as the radiation utilized in wireless
telegraphy, the so-called "radiant heat," and the large class of
radiations possessing varying properties, and known as x-rays.
In present theories the frequency of the vibration (or what is
practically its reciprocal, the wave-length) is considered suffi-
cient to supply the identification, although it seems probable to
the writer that under some conditions, the wave form of the
vibration may be as important as the frequency. Omitting this
point, the zone covered by visible electromagnetic radiations is
comparatively narrow, comprising not more than one octave out
of 45 of the entire section between the longest electric wave, and
the shortest, ultra-violet wave. In other words, only those vibra-
tions which fall between 400 X 1012 and 800 X 1012 cycles per
second are visible, assuming the waves to be simple harmonics.
Whenever an electric charge has its rate of motion changed
the ether is disturbed, a certain amount of energy being trans-
ferred or radiated from the body. In the case in which the
charged particle is accelerated for a time dt with an accelera-
e2a*
tion of "a" the energy radiated is E = fi -==- dt, where e rep-
resents the charge, and V the velocity of light.* Therefore,
dividing by the time during which the change takes place, the
... . , . 2e a
rate at which energy is emitted is — — .
3V
Light production is always associated with matter and usually
matter heated to a high temperature. It is evident from the
equation of the rate of energy emission that since all values
must be divided by V, the velocity of light 2.7 X io10 cm. per
second either enormously large charges or very high accelerations
must be employed to produce any appreciable amount of radia-
tion. Practically, for mechanical reasons, it is impossible to
accelerate charged bodies of a size which can be handled, with
sufficient rapidity to produce any useful radiation. Therefore,
one must discover, if possible, some smaller units which have
a higher ratio of charge to mass, than do "material" objects.
If the case of bodies moving in simple harmonic motion,
* J. J. Thompson, Electricity and Magnetism.
402 TRANSACTIONS I. E. S. PART II
rotation, for instance, be considered, it will be found that the
acceleration varies as the square of the velocity of rotation, and
consequently the equation of energy radiation becomes E = — —
K per cycle where w = the speed of rotation, and K is constant.
Also since the acceleration passes through zero twice each revo-
lution, it follows that the frequency of the emitted radiation
is w.
It is known that for visible radiation w must lie between
400 X 1012 and 800 X io12, the color of the radiated light
depending upon the frequency.
It also follows that as E = — ^7- K = the energy radiated per
3V
„ teeWKdt L . ,. . , , .-
cycle, E* = 1 ^ — = total energy radiated, and that
J 3V
if there are n charges being accelerated the total energy
from all charges will be E/ = 2I — — or making the constants
equal to C, E, = CXMndt.
It is of interest to note in passing, the consistency of this
formula with that which covers the radiation of energy from a
heated body (Stefan's Law, which is based on experiment)
viz., E = C T4 where T = the absolute temperature, in that the
frequency of rotation which varies directly as the temperature
occurs as the fourth power.
The modern conception of matter assumes that the atom con-
sists of a central positive charge (which may or may not be
centered upon a material point) and a number of negative
charges (or electrons) which are in rapid motion around the
central charge. The charges are all units and equal, the differ-
ent properties of matter being derived by the grouping and
numbers of the charges, and by the groupings of the atoms and
molecules.
Since then in all matter there is an unlimited number of
these groups of electrical charges moving about the central
charge of opposite sign, the outer charges being in rapid motion,
DARKAH : SOME CONSIDERATIONS OF LIGHT PRODUCTION 403
one should expect, due to the motion of the charges, a continuous
radiation of electromagnetic waves from matter under normal
conditions. This is found in all bodies and takes place as the
so-called "heat radiation" in accordance with Wein's law, but as
all surrounding bodies are also radiating and conducting heat to
all other bodies, the internal energy of the system is not de-
pleted.
If, however, the temperature of a body be increased the ac-
celeration of the electrons will be increased; and thus also the
energy radiated and the frequency of the radiation will be in-
creased, until at length those frequencies which lie in the
visible spectrum are reached and light produced. The wave-
lengths of the radiations so produced will cover the entire range
of electromagnetic waves since the apparent temperature is an
average of the velocities of each atom, and the average or ap-
parent color will, of course, depend on the average temperature.
If the temperature is raised until the substance is vaporized,
in other words, until comparatively unrestrained motion can
take place among the electrons, a condition resembling resonance
is reached in which the greater part of the radiation of a portion
of the atoms is confined to a definite group of lines. This is
the characteristic of the spectrum observed when sodium, po-
tassium, calcium, salts, etc., are heated in a Bunsen burner. The
effect of this apparent selective action is to absorb energy of all
wave-lengths and to transform a considerable portion of this
absorbed energy into vibrations of a definite wave-length. This
is what is known as selective radiation.
Returning to considerations of the structure of the atom, it
has been demonstrated* mathematically that assuming a given
number of electrons in each atom moving under definite condi-
tions they must take definite configurations. Since varying the
temperatures varies the speed with which the electrons move, and
therefore causes different configurations which are due to dif-
ferent rates of rotation, one should expect the radiant spectrum
to vary with the frequency of resonance for the charged particles.
An interesting converse effect of this is the phenomenon of
markedly increasing the motion of the electrons of an atom by
* N.-Bore and Albert Crebore, Philosophical Magazine, July, 1913.
404 TRANSACTIONS I. E. S. — PART II
the action of light of a definite wave-length. This is applied
practically in photography where chemical reactions are caused
by the acceleration produced by light on the electrons of the
silver atom. The photo-electric effect where electrons are
actually torn from the mass of a conductor in a vacuum by the
influence of light of a frequency in resonance with the motion
of the electrons is another example of the action of light upon
the electrons within the atom. The effect of temperature to
raise the rate of rotation of the electron sufficiently to produce
resonance with the exciting wave is illustrated in those photo-
electric cells which are insensitive to light at ordinary tempera-
tures, but which become exceedingly sensitive at an elevated
temperature.
The theoretically ideal illuminant is one which approximates
daylight, and which produces a spectrum of the frequencies and
quantities necessary to approximate daylight. It is obvious
therefore, that if it is desired to confine the radiation to a limited
range, selective radiation must be relied upon. Since in non-
selective or temperature radiation, due to the constrained condi-
tions surrounding the motion of the radiators as a result of their
numbers and proximity, the vibrations cover all frequencies
and therefore the quantity of energy radiated in this way at any
one frequency will depend upon the number of particles moving
at the necessary frequency. Since there will be relatively few
atoms at the extreme temperatures, there will be relatively little
radiation at the extreme wave lengths. The greater part of the
radiation takes place at the average frequencies, and a curve
plotted between wave length and energy emitted will resemble
the probability curve.
In choosing a measure of the efficiency with which the energy
supplied to a body is converted into useful visible radiation, the
assumption is made that all energy supplied is radiated in some
form. The efficiency, therefore, becomes the ratio of the total
energy radiation to the energy radiated within the range of
visibility, and is used here in that sense.
A consideration of the limited efficiencies theoretically possible
from incandescence or black body radiation alone will serve to
emphasize the advantages of selective radiation. Thus assum-
DARRAH : SOME CONSIDERATIONS OF LIGHT PRODUCTION 405
ing a frequency of 500 X JO12 as an average for normal visi-
bility and taking Wein's Law as a basis, the energy at fre-
quency w, =Ew = -^3- (0 , where the letters have the
same significance as previously except that C is a constant equal
to about 14,395 and e = the base of the natural logarithm system.
This gives approximately 6,ooo° absolute as the necessary
temperature for maximum efficiency of radiation from an incan-
descent body. The efficiency at this temperature while theo-
retically about 10 per cent, cannot be approached in actual prac-
tice since the highest known temperature (the vaporization point
of carbon) is about 3,800° At this temperature the efficiency
of radiation of light by incandescence is about 5 per cent, while
at the melting point of tungsten the efficiency is about 3 per cent.
It is evident, therefore, that unless a conducting material is
discovered having higher melting point than carbon, luminous
efficiencies from incandescent bodies must remain limited. At
the present time it is, of course, the vaporization of the filament
of an incandescent lamp which limits the safe temperature, but
even when this difficulty has been removed as seems commer-
cially possible in some cases by maintaining a high pressure
around the filament, the low melting point becomes a serious
limitation for which no absolute remedy is in sight.
The limitations of incandescent illuminants may be changed,
of course, in case a substance having a radiant spectrum lying
very largely within the visible zone were discovered — in other
words, one exhibiting selective radiation in a high degree while
solid.
Referring again to the equation previously derived for the
total quantity of energy radiated at any given frequency, —
E = \Ce'7t>*ndt, and assuming, what is generally accepted,*
that e is the charge upon an electron, the only variables are the
frequency, the time and number of electrons involved. It fol-
lows therefore, from the expression that the maximum intensity
or maximum energy radiated by each unit will be obtained with
* J. J. Thompson, Conduction of Electricity through Gases.
3
406 TRANSACTIONS I. £. S. — PART II
the highest frequency and when the greatest possible number of
electrons are involved.
In solid matter the particles are too closely associated to vi-
brate without considerable restraint, it is evident, therefore, that
to secure the effect of resonance, or selective radiation, the mov-
ing particles should be as far separated as possible. Therefore,
in most cases, selective radiation is confined to gases or vapors,
and the study of radiation efficiencies becomes the study of gas
radiation.
From previously quoted formula, it will be noted that the
quantity of light radiation varies directly with the number of
electrons which are subjected to the accelerating force. There-
fore, to secure the maximum intensity per unit volume of radia-
ting gas, the pressure should be as high as possible, as this means
the presence of a large number of electron groups. The mere
presence of a large quantity of electrons, however, is not in
itself an insurance of either a high intensity of light or a high
efficiency. It is only those electrons which are subjected to a
change of rate of motion (or acceleration) between certain limits
which are effective as radiators. Since the acceleration of an
electron sufficiently to radiate in the wave lengths included in
the limits of visibility implies the acceleration of the adjacent
systems to nearly the same degree, a rough measure of the effi-
ciency of light production in a gas, will be the ratio of the num-
ber of radiating systems to the total number of such systems
being acted upon. To secure a high efficiency, it is therefore
desirable to use as small a quantity of gas as possible and to
cause as many of the electron systems as possible to take part
in the change in rate or direction of motion. In passing, it is
of interest to note that the electric arc approaches this condition
more nearly than other illuminants as the pressure is compara-
tively high and the percentage of active particles relatively large.
A chemical reaction between gases is the best known method
of accelerating a large percentage of the atoms present. The
percentage acted upon will, of course, be definitely determined by
the phase rule, and will depend upon the rate of dissociation and
of recombination at the temperatures at which the gas is worked.
It is of passing interest to note that in the various types of flame
DARRAH : SOME CONSIDERATIONS OF EIGHT PRODUCTION 407
carbon arcs and metallic flame arcs advantage is taken of chem-
ical combination to secure a high intensity, and a high efficiency.
In these arcs the various cerium, titanium, or calcium compounds
are heated to the temperatures at which dissociation takes place
while the rapid movement of the gases carries the dissociated
products to a region of lower temperatures where recombina-
tion takes place. Thus probably the majority of the constituent
compounds undergo a chemical reaction at least twice which re-
sults in a considerable acceleration of the electrons, and there-
fore, in light production. The striking difference in luminosity
of different portions of the flame arcs, and their sheath-like
appearance due to the varying intensity of the light in the sur-
rounding layers is a confirmation of this theory, the inner very
hot layer being the source of relatively little radiation while the
surrounding sheath which is the seat of the dissociation and
recombination of the chemicals of the arc is the source of most
of the light.
It seems probable that the high efficiencies of the Welsbach
mantle are explainable on the same basis, namely that at the
high temperatures of the gas flame, an unstable compound of
thorium and cerium with oxygen or other elements is formed
and that the continual formation and decomposition of this
compound involves many more atoms (and thus effects many
more electrons) than would be possible by merely heating the
mantle. This is in part substantiated by the fact that heating
the mantle in an inert atmosphere does not produce very much
light. In passing it is of interest to note that the addition of
thorium (which appears to be essential to the successful opera-
tion of the Welsbach mantle) does not appreciably improve the
efficiency or intensity of the light from a flame carbon arc in
which cerium is the chief illuminant. A possible explanation
of this fact is that the function of the thorium is to lower the
temperature at which the unstable compounds form, while in
the flame carbon arc the temperature is sufficiently high to make
this unnecessary and thus removes the advantage of thorium.
In addition to considerations of the intensities of the radia-
tion, the position and number of the lines in the spectrum of a
substance is of great importance in determining its value as a
408 TRANSACTIONS I. E. S. — PART II
source of light. Quite recently some convincing theories have
been developed which account for the position of the spectrum
lines very consistently. These theories assume that the atoms
of the various elements are capable of existing in a number of
more or less stable states due to the rate of motion and the
number of the component electrons. In passing from one con-
dition of stability to the next stable condition a certain amount
of radiation is emitted which will have a definite wave length
due to the rate of acceleration of the electrons at that instant.
The various stable states can actually be demonstrated by models
in which small charged spheres are arranged around a central
charged sphere of opposite sign.
A mathematical analysis of the behavior of charged particles
moving about a central charged particle of opposite sign results in
a formula from which the calculation of the exact position of the
spectrum lines can be made in the case of hydrogen and some
of the simpler elements.* This theory leads to the supposition
that chemical affinity is the result of electrical forces between
the groups of electrons and that at the instant a change from
one stable state to another stable state occurs (or at resonance)
the chemical compound is destroyed. On this basis it would
seem that those elements which have the greatest number of
transition stages (and therefore, have the most intense and com-
plete spectrum) are reluctant to combine with other elements and
would most readily dissociate when combined. This is con-
firmed by experiment in that hydrogen, nitrogen, helium, argon
and neon, etc. (the elements which are perhaps the least active)
are peculiarly efficient as illuminants in a Geissler tube or its
equivalent, while oxygen, flourine, chlorine and the so-called
more active elements are relatively very inefficient as radiators
and have relatively very few bright lines in their spectrum.
SUMMARY.
Summarizing the theories which have been advanced, it ap-
pears that we cannot hope to secure a radiation efficiency from
known incandescent solids by simple black body radiation, which
will be much greater than 5 per cent., due to the low melting
points, and that therefore, the greatest expectation of obtaining
* A. Crehore, Philosophical Magazine, July, 1913.
DARRAH : SOME CONSIDERATIONS OF EIGHT PRODUCTION 409
a high efficiency lies along the lines of selective radiation from
electrified or incandescent gases. Since the highest efficiencies
will be obtained by using the minimum amount of gas and forc-
ing the greatest possible number of electron systems to take part
in the acceleration, some form of arc (or gaseous conductor)
appears to be the most promising field for the future. Further,
as different materials have different radiant efficiencies, the arc
should be fed with materials which will give the maximum radi-
ation within the limits of visibility, and in the proportions which
produce white light. In selecting the material for supplying
the radiation of an arc it should be borne in mind that chemical
inertness is frequently an indication of an extended spectrum of
considerable intensity.
CONCLUSIONS.
The conclusions which the facts summarized in this paper
seem to indicate are, that in spite of the remarkable work which
has been done, in raising the efficiency of the incandescent lamp,
it is nearing its maximum theoretical efficiency. Although the
fact should be kept in mind that ease of application, size of units,
maintenance cost and first investment, may have as much weight
in deciding upon the type of illuminant used as the absolute
efficiency and, therefore, illuminants employing incandescent
solids will undoubtedly have a place in the art for a considerable
period. However, the vapor conductor type of illuminant, as
for instance the flame arc and metallic arc, has only begun to
approximate its possible efficiencies and therefore, has a very
large field ahead of it.
4IO TRANSACTIONS I. E. S. — PART II
THE PENTANE LAMP AS A WORKING STANDARD.*
BY E. C. CRITTENDEN AND A. H. TAYLOR.
Synopsis: This paper recommends the use of tested pentane lamps
as secondary standards of candle-power when electric standards are not
available, and gives a detailed statement of the method of testing followed
at the Bureau of Standards, with general directions for the use of the
lamps. The effects of variation in pentane and in atmospheric conditions
are discussed, a correction for the former is proposed, new determina-
tions of the humidity correction factor are given, and a chart is provided
to facilitate the reduction of observations to normal candle-power.
INTRODUCTION.
In the Transactions of the Illuminating Engineering Society
there have already been published two papers1 dealing with the
work on flame standards of candle-power which has been done
at the Bureau of Standards. These papers were intended to be
particularly reports of progress in the experimental work which
has been done on various types of lamps. The investigation of
the lamps was made with the double purpose of determining
whether any of them were capable of furnishing a check on a
possible drift in the value of the unit now maintained by electric
incandescent standards, and of finding a satisfactory working
standard for use where electric standards are impracticable.
The results of that work proved to be distinctly favorable to
the use of the Harcourt io-candle pentane lamp as a secondary
standard. Since the publication of the papers mentioned the use
of such lamps has become much more extensive, and the present
paper is written with a somewhat different object in view, namely,
to cover the questions which have arisen regarding their opera-
tion. This involves a fulness of detail which will not be of much
interest to the general reader, but which it nevertheless appears
desirable to put in print because some of this information is not
generally available.
* A paper read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
1 Transactions, Illuminating Engineering Society, Vol. V, pp. 753-778, 1910 ; and
Vol. VI, pp. 417-432 ; 1911.
CRITTENDEN AND TAYLOR: PENTANE LAMP 4II
Extent of Use of Pentane Lamps. — The number of pentane
lamps tested by the Bureau is not an exact indication of the
extent to which the use of the lamps has increased, because many
have been standardized in other laboratories and many others,
especially in the early years of their use, were not standardized
at all. Nevertheless the number tested, as given in the table
below, is an indication of the widespread adoption of this
standard.
Pentane Lamps Tested by the Bureau of Standards.
1908 4
1909 6
1910 7
191 1 17
I912 35
1913 (6 months) 15
Total 84
Since in many cases one of these lamps furnishes the basis on
which the quality of the gas supply of a city, so far as candle-
power is concerned, is judged, the importance of the pentane
lamp in present practise will be appreciated.
POSSIBLE USES OF PENTANE LAMPS.
While the lamp has found its widest use in gas testing, where
a flame standard is preferred because the effects of atmospheric
conditions on it compensate for similar effects on the gas flame,
its usefulness is by no means limited to work of that kind. The
corrections to determine the actual candle-power are easily made,
and the lamp affords a fairly convenient method of obtaining a
basic standard for laboratories which lack the batteries and accu-
rate instruments which are necessary to get reliable values from
electric standards.
In some cases the pentane lamp may thus serve as a basis of
standardization even when it may be found more convenient to
use other standards for the actual tests. For example, when a
series of measurements are made extending over several hours,
another source such as a kerosene lamp, an Edgerton standard
or a gas mantle may be used as a constant comparison light at
one end of the bar, its value being fixed by substituting a pentane
lamp for the test flame or lamp at the beginning and the end of
the series. An electric lamp kept at constant current may be
412 TRANSACTIONS I. E- S. — PART II
used in the same way as a comparison light. This is exactly what
is done in most portable photometers, and the calibration of such
photometers can be based on a pentane lamp as well as on an
electric standard. A modification of this practise which is useful
where tests for approximate values have to be made at a number
of stations is to provide for each station a portable standard unit
consisting of a low voltage lamp and a storage battery, the whole
unit being brought in at regular intervals for recharging and
recalibrating against the pentane standard. From a curve show-
ing the variation in candle-power with time of use for such a
unit one can obtain fairly satisfactory values without the use of
either regulating resistances or electrical measuring instruments.
The details of such a system in actual operation are given by
Mr. H. h. Farrar, in the Gas Age of April 15, 1913 (pp. 407-
409). In using such standards, as in any case where electric lamps
are compared with flames, the effects of humidity are important.
The readings can, however, be easily corrected for this by the use
of such a chart as is given in Fig. 3.
It is hardly necessary to give here a long discussion of the
relative advantages of different lamps, and of the reasons which
are leading to the common use of the pentane lamp as a working
standard in this country. There are other lamps which may be
used in cases where their value can be frequently checked or
where the accuracy required is not great. In some other cases,
especially where portability is an important consideration, the
Hefner lamp is to be recommended; by making many measure-
ments under favorable conditions very good results can be ob-
tained with it, but in general its low candle-power, comparatively
red light, extreme sensitiveness to drafts, and unsteadiness at
temperatures above 70 deg. F. make it difficult to use. The pen-
tane lamp has the disadvantages of being large and not easily
portable, of using fuel which is expensive and somewhat danger-
ous, and of requiring more air than ordinary ventilation will
supply. In spite of these faults its use is increasing, chiefly
because it appears to be the only standard of candle-power now
in use, other than incandescent electric lamps, which can be
relied upon to give under the usual working conditions the degree
of accuracy expected in present commercial practise.
CRITTENDEN AND TAYLOR: PENTANE LAMP 413
STANDARDIZATION OF LAMPS AT BUREAU OF
STANDARDS.
NECESSITY OF PHOTOMETER TEST.
The necessity for standardization of each lamp by a photo-
metric test arises from the fact that different lamps, even of the
same maker and supposedly of the same type, show differences
in candle-power which are sometimes as great as 4 per cent.
The construction is such as to make it difficult to determine after
a lamp is assembled whether it is built exactly according to
specifications. In fact, exact adherence to specified dimensions
might not remove the differences between lamps unless other
conditions also were definitely specified. For example, the cast-
ings used in the American lamps have a rough inner surface ;
this roughness probably facilitates the transfer of heat to or from
the air, and it may also affect slightly the flow of air through
the passages. Whatever may be the cause of the differences
between lamps, the candle-power of the individual lamp is the
important thing, and hence the logical test to make is a deter-
mination of the candle-power. In lamps tested at the Bureau
details of construction are not given special attention unless some
part is so made that it appears to be a possible cause of varia-
tion in the candle-power of the lamp.
Since the lamp itself is only an instrument for producing the
actual standard, the flame, there is little significance in a value
given for the lamp unless other factors, such as air and fuel
and other conditions of operation, are known. The value cer-
tified for a lamp applies strictly only for its use under conditions
identical with those under which it was standardized ; it may
therefore be worth while to particularize rather fully the method
of testing followed at the Bureau.
PHOTOMETER ROOM.
The photometer room used hitherto is approximately 26 x 18 x
12 feet (7.92 x 5.49 x 3.66 m.). Ventilation is obtained by a
tempered air heating system. The ordinary ventilation would not
keep the air sufficiently pure if the flame gases were allowed to
escape into the room, and consequently above the lamp is hung a
hood 2 feet (0.6 m.) in diameter with an outlet into one of the
ventilating ducts. The draft into this hood is just strong enough
414 TRANSACTIONS I. E. S. — PART II
to ensure that all the gases rising from the lamp shall pass into
the hood and thus be carried out of the room. This arrange-
ment is so effective that no perceptible change in the candle-
power of the lamps due to vitiation of the air occurs in several
hours continual burning. In regular work, however, the room
is usually aired out at least once every hour.
The photometer room is in a corner of the building and often
has troublesome drafts, in consequence of which it has been found
that the performance of the lamps has been improved by screen-
ing them on three sides and sometimes on all four sides. This
is done by a wooden frame-work about 3 feet (0.91 m.) long,
2 feet (0.61 m.) wide and 3 feet (0.91 m.) high, carrying a strip
of cloth which extends from the top of the frame down to 6
inches (15.24 cm.) from the bottom, thus allowing free access of
the air from all sides below the lamp, but preventing any drafts
from striking the flame or the circulatory system of the lamp
directly. This screening was adopted after many trials of vari-
ous forms of enclosure had shown that in most of them the lamps
gave values slightly different from those obtained in the open,
even when no vitiation of the air in the enclosure could be
detected.
In special tests of the state of the air a Zeiss refractometer has
been used to determine the quantity of C02 present. Measure-
ments of atmospheric moisture are regularly made with 2
Assmann psychrometers, and of pressure by a mercury barometer.
The details of the calculation of the moisture will be given later
in connection with the corresponding corrections.
The photometer outfit is of the standard form regularly used
at the Bureau, with Lummer-Brodhun contrast head and a re- '
cording device which has been described in the Bulletin of the
Bureau.2 The substitution method with constant intensity on
the disk is used as in all standard work, that is, a constant "com-
parison lamp" is connected to the photometer head at a fixed
distance from it, while the electric "working standards" and the
pentane lamps under test are alternately put in on the other side
for reading.
- G. W. Middlekauff, Bulletin B. S., 7, p. iS, 1910. Reprint No. 144, and Electrical
World, Vol. IvVI, p. 153, (July 21) 1910.
CRITTENDEN AND TAYLOR: PENTANE LAMP 415
ADJUSTMENT OF LAMPS.
Before putting the lamp on the photometer the chimney is
inspected by looking down through it to see that it is set centrally
over the burner, and each time the lamp is lighted the gauge is
applied to make sure that the chimney is at the proper height,
that is, 47 mm. above the burner when cold. The chimney seldom
has to be reset, but occasionally one creeps slightly with heating
and cooling.
In setting up the lamp on the table care is taken to level it so
that the chimney shall be vertical. The levels (plumb bobs on
later lamps) attached for this purpose are not always exact, and
this leveling is done by dropping a plumb line from the center
of the top of the chimney so that the bob should hang centrally
in the burner. The height of the lamp is so adjusted that the
middle of the flame is on a level with the center of the photometer
disk. The directions of the Gas Referees are such as to bring
the bottom of the chimney on a level with the center of the disk,
and this setting makes the illumination on the disk follow the
inverse square law more exactly than if the lamp is higher. The
American Gas Institute's Committee on Taking Candle-power of
Gas has, however, recommended the other position and in order
to avoid confusion the Bureau has followed this recommenda-
tion. The lamp is usually placed so that the line of the bar passes
through the lamp standard as well as through the center of the
burner and chimney. In some cases, however, on request, the
lamps are turned 90 deg. so that a plane passing through the
chimney and the standard is at right angles to the line of the
bar. No difference has been detected in the values obtained in
the two positions for any lamp. In either case the chimney is so
turned that the light from the mica window does not fall on the
photometer disk or on the lamp standard or the pentane feed
tube.
In measuring distances from the lamp the center of the
burner is taken as the point of departure. The photometric ob-
servations are made at a distance of approximately 1 meter. For
setting the lamp at the proper point on the scale it has been
found convenient to follow the English method of using a rod
having near one end a cylindrical plug which fits into the burner.
416 TRANSACTIONS I. E. S. — PART II
This rod is made of such length that the other end just touches
the photometer head when the center of the burner is one meter
from the disk. This fixes very exactly the point of the scale
which is i meter from the lamp, and a line on the record sheet
is set at this point. The distance of the comparison lamp is then
so chosen as to make the settings on the pentane lamp fall near
the line, and in working up the sheet the distances of the groups
of points which represent readings on the lamp are measured
from this line.
OPERATION OF LAMPS.
Pentane. — The pentane used is tested with respect to density
and purity; the effects of possible variation in the fuel are so
important that a later section of this paper is devoted particularly
to that question.
Flame Height. — The method of controlling the flame height has
been the subject of considerable difference of opinion. At the
Bureau no difference in candle-power has been found to result
from the use of the various cocks at which regulation may be
effected. Even when the saturator is entirely closed off from the
air, and the pressure of the vapor in it is allowed to run up to
9 inches (29.53 cm.) of water, regulation of the flame height
being made by a cock near the burner, no departure as great as
1 per cent, from the normal candle-power occurs. In standard-
izing the older lamps, in which such cocks are provided, they have
been used. The newer lamps do not have these cocks, and flame
height is controlled by the outlet cock of the saturator, in which
case greater ease of adjustment is obtained by clamping a lever
on to the handle of the cock.
Another question concerning which there are differences of
opinion is that of the proper flame height. In English lamps
the top of the flame should be set at the middle of the lower
mica window, about 27 mm. above the bottom of the chimney.
The lamps are so proportioned that this height of flame gives a
maximum candle-power and a small shift up or down makes very
little change in candle-power. This is shown in curve E of
Fig. 1 where abscissas are candle-powers and ordinates are
heights of flame in the chimney. The effect of personal error in
judgment of height of flame is thus made very small.
CRITTENDEN AND TAYLOR: PENTANE LAMP
4*7
The American makers have placed the crossbar of the window
lower, and direct that the flame be set with the tips at the top
of the bar, which is about 22 mm. above the bottom of the
chimney. As is shown in curve A of Fig. 1 this is not high
enough to give the maximum candle-power, and at this height
small changes cause relatively large variation of candle-power.
At the Bureau it has been considered important to retain the
advantage given by using the maximum candle-power, and inci-
dentally to adhere to the original manner of operating the lamps.
The practise has therefore been definitely adopted of first de-
termining the height of flame which gives the maximum candle-
'
1 AAMERiCAN LAf
EEN6USH LAMP
P
»«»«.
L.
mxm
jj
H,
\ \
\
\ 1
/ i
r
___.
/ /
""""
■OTTO.*
// "
fT?
*
'
\
CANDLE-POWER
Fig. 1.— Variation of candle-power with pentane lamp, with height of flame.
power for a given lamp, and then standardizing the lamp with the
flame at that height. With American lamps the maximum is
usually obtained when the top of the flame is about a centimeter
above the bar. A line is marked on the window to show the
height used, and the fact is stated in the certificate furnished with
the lamp.
Time of Burning. — The necessity of waiting a sufficient time
after lighting the lamps before making measurements was em-
phasized in the earlier papers where curves were given showing
the variation of candle-power with time after lighting for English
and American lamps. It is important to note that the candle-
power rises rapidly at first, going above the normal, and then
settles back to a fairly constant value. Considerable attention
4l8 TRANSACTIONS I. E. S. — PART II
has been given to the measurement of the variations in tempera-
ture in various parts of the lamp which accompany these changes
in candle-power. The details of the measurements will not
be given here, but it may be said that they have strengthened the
conviction that most of the change in candle-power caused by
variation in conditions of operation may be attributed to the
air circulating system and the variations in the flow of air which
arise from changes in the relative temperatures of parts of the
lamp. There is a slow change of temperature for a considerably
longer time, but the candle-power is practically constant after
30 minutes in the case of the old style American lamps and after
20 minutes with the newer lamps. The excess of the maximum
over the final candle-power is about 3 per cent, and 2 per cent,
respectively for the two types. Lamps of the English type run
from 1 to iy2 per cent, high at the maximum and may be con-
sidered constant after 15 minutes burning. The smaller heat
capacity of these lighter lamps, which allows them to reach a
steady state sooner, seems also to make them more susceptible
to variations caused by changing drafts.
BASIS OF STANDARDIZATION.
The fundamental unit in terms of which lamps are certified
is the international candle3 as maintained by the primary elec-
tric standards of the Bureau. The standards actually used in
tests are a group of 7 carbon lamps operated at low voltage so
as to match the pentane in color. Since the flame is much redder
than the ordinary electric standards the calibration of these spec-
ial standards has taken considerable labor to insure that no
important error be introduced because of this color difference.
The group has been compared with the Bureau's regular working
standards in three series of measurements with a number of
different observers, under the direction of Dr. Middlekauff, and
the average values have been as follows : —
Candle-power
I91I 498
1912 4.98
1913 4-97
It appears therefore that the uncertainty in the value of these
3 Bureau of Standards, Circular No. 15.
CRITTENDEN AND TAYLOR: PENTANE LAMP 419
standards cannot be very great. As a check on their permanence
another group carefully compared with them is preserved and
used only for checking the working standards. A similar group
has also been prepared to be sent to England in order to obtain
a direct comparison between the standards of the National
Physical Laboratory and those of the Bureau at pentane color.
To obtain the normal candle-power of the flame, corrections
are made for the effects of atmospheric moisture and pressure.
These corrections will be considered in a separate section since
they must be used whenever it is desired to obtain absolute candle-
power values with a flame standard. As a check against the
introduction of errors by other conditions, such as poor ventila-
tion, which might affect the flames but not the electric standards,
a pentane lamp of known candle-power is always included
in tests.
NUMBER OF MEASUREMENTS.
The number of measurements made on each test lamp in
the course of standardization depends somewhat on the con-
sistency of the results obtained. Usually each lamp is placed on
the photometer about six times, and each time four groups of
sets are made, each group consisting of 25 to 40 settings of the
photometer. The candle-power result for each group is worked
up separately, and if the average deviation of these values from
the final mean is not materially greater than one-half per cent.,
and the values obtained for the check lamp run in the test are
also normal, the test is considered satisfactory. If these condi-
tions are not fulfilled, further measurements are made until it
is believed that the mean candle-power of the lamp is sufficiently
well determined. The candle-power is certified to the nearest
tenth of a candle unless the average of test results falls very
nearly half way between tenths, in which case a subscript 5 is
given in the hundredths place; this is written as a subscript to
indicate that it is not considered as definitely established, but
merely as representing the average of test results.
ACCURACY OF VALUES.
With all the precautions taken it is believed that the values
certified for the lamp are correct within one per cent., that is,
420 TRANSACTIONS I. E. S. — PART II
that under similar conditions the lamps should give average
results within one per cent, of those certified. It has occasionally
happened that a second test on a lamp has given a result fully-
one per cent, different from the value originally certified. The
result obtained from a single time on the photometer sometimes
departs as much as 2 per cent, from the mean, but the maximum
deviation is seldom as great as this. As for the permanence of
the calibration no evidence has been obtained indicating any
appreciable change with time. For the Bureau's two Chance
lamps which are regularly used as checks on tests the average
candle-powers found in 1910 were 9.87 and 9.89; the averages
for the first six months of 1913 have been 9.90 and 9.87.
Conditions in different laboratories are not likely to be exactly
the same, and in order to depend on reproducing values in dif-
ferent places as closely as one per cent, one would probably need
to give some care to reproducing conditions of operation. It
would seem, however, that under any reasonably good conditions
the difference ought to be well within 2 per cent., for, using good
pentane and correcting for atmospheric moisture and pres-
sure, it is difficult to produce that much variation in candle-
power by any intentional change of conditions except vitiation of
the air or incorrect flame height.
GENERAL DIRECTIONS FOR USE OF PENTANE
LAMPS.
It is hardly practicable to give here detailed directions as to
the exact procedure to be followed in any case, but attention may
be called in a general way to some precautions which should be,
taken in using the lamps. Many of the details of adjustment
and operation which have already been discussed will be merely
mentioned here. The best results will presumably be obtained
by following the same methods in operating a lamp as were used
in standardizing it. If the lamp to be used has been standardized
at the Bureau a reference to the preceding pages will answer
most questions as to proper procedure. If the value of the lamp
has been assigned by another laboratory such questions should
usually be referred to that laboratory.
CRITTENDEN AND TAYLOR: PENTANE LAMP 421
VENTILATION AND EFFECTS OF VITIATION OF AIR.
Since all flames depend upon combustion it is to be expected
that their intensity will vary with the proportion of oxygen and
of other constituents in the air supplied to the gaseous fuel of the
flame. To maintain a flame in a constant condition requires,
therefore, not only uniform fuel, but also a uniform proportion
of oxygen in the air supplied to the flame. Good ventilation is
consequently desirable for all work with flames, and is indis-
pensible where lamps of other kinds are to be compared with a
flame standard. The quantity of air which must be supplied de-
pends largely on the size and number of flames in the room and
also on the method of ventilating. The flames themselves set up
currents of air, and if possible these currents should be utilized
to remove immediately from the room all the vitiated air coming
from the lamps. A hood placed above the photometer with a
rising pipe to lead the warm air out of the room is most effective.
If the gases from the flames are allowed to diffuse into the room
and are simply diluted by the air entering, a much larger amount
of fresh air will be needed. In such a case it has been esti-
mated4 that to keep the carbon dioxid content of the air down to
six parts in 10,000, 3,000 cubic feet (84.95 cu. m.) of fresh
air per hour must be supplied for each person in a room.
A committee of the American Gas Institute5 has on this
basis estimated that 36,900 cubic feet (1045 cu. m.) of air per
hour should be supplied for a photometer room where two
persons work with a pentane lamp and a 5-foot (141.6 1.) gas
flame, since in the production of carbon dioxid the gas flame is
approximately equivalent to 6 persons and the pentane lamp to
four.
In a room where air enters at the bottom and escapes at or
near the top, the quantity estimated as above (about 12,000 cubic
feet (340 cu. m.) per hour for a pentane lamp) is probably
ample for any purpose, and if care is taken to allow direct escape
of the flame gases a much smaller amount is sufficient.
Some caution is necessary in making calculations based on the
quantity of carbon dioxid in the air. Experimental determina-
4 Kent, Mech. Eng. Pocketbook ; 1912 Ed., p. 654.
6 Proceeding! A met ican Gas Institute, Vol. II, pp. 481-4S2, 1907.
4
422 TRANSACTIONS I. E. S. — PART II
tions of the effect on flames have sometimes been made by adding
carbon dioxid to the air, and it must be remembered that the
dilution of the air by a quantity of added C02 has relatively little
effect as compared with the conditions arising when the same
amount is formed in the process of combustion, using up the
oxygen of the air. In the formation of five cubic feet of C02
by the combustion of pentane, for example, 8 cubic feet of
oxygen are used. Since the abstraction of i cubic foot of
oxygen has practically the same effect on the composition of the
air as the addition of 5 cubic feet of inert gas, it is evident that
the reduction in the amount of the active oxygen is of much
greater importance than the increase in the diluting gases. The
effect produced on the Hefner lamp by adding one part of C02
to 1,000 of air, according to Liebenthal,6 is a reduction of 0.7 per
cent, in the candle-power, but the generation of the same propor-
tion of C02 by combustion and breathing in a room has been
found7 to be accompanied by a decrease of 2.2 to 3 per cent, in
the Hefner, the pentane lamp and other flames. In general the
effect depends on the manner of production of the C02, since
the greater part of the decrease is due not to the presence of C02
but to a deficiency of oxygen. Determinations of the amount of
C02 in the air consequently do not furnish sufficient data for exact
correction for vitiation of the air, but are useful in that they
enable one to judge the effectiveness of the ventilation.
In many cases, however, determinations of C02 are not prac-
ticable and the effectiveness of the ventilation must be judged
by other means. When facilities are available for setting an
electric lamp repeatedly to the same current a direct test can be
obtained by making a series of measurements of the electric lamp
against the pentane and finding whether the latter shows a grad-
ual decrease after it should have reached a constant value.
Another test which is as definite as determination of the carbon
dioxid and far more easily carried out is afforded by careful
measurement of the humidity, for the processes which use up
* Zeitsch.f. Ivstrumentenkunde, Vol. XV, p. 157, 1895.
" C C. Paterson, Collected Researches, National Phys. Lab., Vol. Ill, p. 49, 1908.
Butterfield, Haldane & Trotter, Journal Gas Lighting, Vol. CXV, p. 290, 1911, and
American Gas Light Journal, Vol. XCV, p. 145, 1911. Also unpublished tests of Bureau of
Standards.
CRITTENDEN AND TAYLOR: PENTANE LAMP 423
oxygen add water as well as C02 to the air. If the water vapor
regularly increases by an appreciable amount during the opera-
tion of the lamp the ventilation is not satisfactory. An increase
of one liter of water vapor per cubic meter of air, when caused
by poor ventilation, is quite regularly accompanied by such vitia-
tion as to cause about 2 per cent, decrease in the candle-power
of flames, in addition to the 0.6 per cent, decrease caused by the
water vapor itself.
In designing ventilating inlets ample capacity should be pro-
vided to allow a slow flow of air into the room in order to avoid
drafts which would cause unsteadiness of the flame. Similarly
if a hood is used as recommended the outlet must be so arranged
that the flow of air into the hood shall not be too vigorous.
Troublesome drafts are apt to arise if the walls of the photometer
room differ much in temperature from the air. It is therefore
desirable that none of them shall be exterior walls of the building.
If such walls are unavoidable they should be either jacketed
with non-conducting material or covered by a false wall with
an air-space. If possible the room should be so free from drafts
that the flame will burn steadily without other protection than
the necessary photometric screens. If further protection from
drafts is necessary a screen of the form described on a preceding
page should be used.
PREPARATION OF LAMP.
In preparing the lamp for use one should observe the following
details : centering of chimney over the burner, height of chimney
above the burner, direction in which the mica window is turned,
amount of pentane in the saturator, height, orientation and level-
ing of the lamp, and its distance from the middle point of the bar
or some other definite point on the scale. It is advisable, at least
in the beginning, to check the last two adjustments by direct
measurement previously described. If then the plumb bobs (or
the level and bob) are found to be correct they may be used
thereafter.
The saturator should be from one-third to two-thirds full of
pentane at starting, and the height of the liquid as seen against
the window of the saturator should never be less than */£ inch
(3.175 mm.). Since pentane is very volatile and inflammable,
424 TRANSACTIONS I. E. S. — PART II
and the heavy vapor flows downward, it is extremely hazardous
to fill a lamp while it is burning. This should never be attempted
under any circumstances, and it should be an inviolable rule of
the laboratory that no pentane in an open vessel be brought near
any flame.
OPERATION.
Immediately before the lamp is lighted the height of the chim-
ney should always be tested with the gauge, and occasionally the
gauge itself should be measured to see that it remains 47 mm.
long. To light the lamp open first the saturator inlet cock; then
holding a lighted match over the burner open the outlet cock
gradually. If the lamp has a regulating cock near the burner,
the saturator cocks may both be opened and the flame controlled
by this regulating cock. If no such cock is provided the saturator
outlet cock is used for regulation. Usually on opening the cocks
the vapor will flow so that the lamp will light, but it is sometimes
necessary to start the flow of vapor by blowing gently into the
saturator inlet. Sometimes also the lamp at first burns with a
small blue flame because the heavy vapor has flowed down into
the burner and prevents the normal circulation of air from start-
ing. If this happens shut off the cocks, and blow up into the
outer chimney to start the air, keeping below the burner to avoid
the small burst of flame which may result if the flame has not
entirely died out in the tip. The lamp will then light normally.
After it is lighted see that the conical hood around the flame is
so placed that the whole flame is visible from the photometer
disk.
The flame should be kept at approximately the correct height,
and no measurements should be made till the lamp has burned
15, 20, or 30 minutes according to whether it is of the English
type or of the new or old American forms. Unless otherwise
specifically stated in the lamp certificate for the American lamps
the proper flame height is obtained when the tips are just above
the crossbar of the window, while in English lamps the flame
should extend half-way up the lower window. As already stated
American lamps standardized recently at the Bureau of Standards
have a line on the window to which the top of the flame is to be
set. After the lamp has burned a few minutes such changes in
CRITTENDEN AND TAYEOR : PENTANE LAMP 425
flame height as occur are gradual, and except in the most careful
work it is not necessary to have a special observer to watch the
flame.
The lamp is extinguished by shutting off the saturator cocks.
When it is not in use both cocks should be kept closed and a cap
should be placed over the burner to prevent injury to it or the
collection of dust in the passages. All parts of the lamp should
be kept well blackened.
PENTANE: PREPARATION, TESTING AND USE.
The directions of the Gas Referees for the preparation and
testing of pentane are as follows :
Preparation. — Light American petroleum, such as is known as Gasoline
and used for making air-gas, is to be further rectified by three distilla-
tions, at 55 deg. C, 50 deg., and 45 deg. in succession. The distillate at
45 deg. is to be shaken up from time to time during two periods of not
less than 3 hours each with one-tenth its bulk of (1) strong sulphuric
acid, (2) solution of caustic soda. After these treatments it is to be
again distilled, and that portion is to be collected for use which comes
over between the temperatures of 25 deg. and 40 deg. It will consist
chiefly of pentane, together with small quantities of lower and higher
homologues whose presence does not affect the light of the lamp.
Testing. — The density of the liquid pentane at 15 deg. C. should not
be less than 0.6235 nor more than 0.626 as compared with that of water
of maximum density. The density of the pentane when gaseous, as com-
pared with that of hydrogen at the same temperature and under the same
pressure, may be taken. This is done most readily and exactly by Gay
Lussac's method, under a pressure of about half an atmosphere and at
temperatures between 25 deg. and 35 deg. The density of gaseous pen-
tane should lie between 36 and 38.
Any admixture with pentane of hydrocarbons belonging to other groups
and having a higher photogenic value, such as benzene or amylene must
be avoided. Their presence may be detected by the following test.
Bring into a stoppered 4-oz. bottle of white glass 10 cc. of nitric acid,
specific gravity 1.32 (made by diluting pure nitric acid with half its bulk
of water) ; add 1 cc. of a dilute solution of potassium permanganate,
containing 0.1 gram of permanganate in 200 cc. Pour into the bottle
50 cc. of the sample of pentane, and shake strongly during five successive
periods of 20 seconds. If no hydrocarbons other than paraffins are
present, the pink color, though somewhat paler, will still be distinct; if
there is an admixture of as much as 1/2 per cent, of amylene or benzene,
the color will have disappeared.
For the benefit of those who are not chemists it should be said
that this last test should always be preceded by a blank test to
426 TRANSACTIONS I. E. S. — PART II
check the purity of the reagents, for if the nitric acid is not pure
it will decolorize the permanganate. The acid should be kept in
the dark to avoid deterioration.
The pentane used at the Bureau is tested with respect to purity
and density of the liquid. The density is conveniently deter-
mind by a hydrometer carrying a thermometer. The correction
for temperature is important, being about 0.00 1 per degree Centi-
grade or 0.00055 Per degree F. In the sealed ether cans
in which pentane is now purchased no difficulty has been found
in keeping a supply which fulfills the specifications when the cans
are opened, but the more rapid evaporation of the lighter frac-
tions raises the density as the fuel is consumed in the lamp. It
is quite impracticable to work with pentane within the limits pre-
scribed, for the density is certain to be too high before one-tenth
of the pentane is consumed; in fact the density usually reaches
0.635 when a little over half has been used.
CORRECTION DETERMINED BY DENSITY.
The directions of the London Gas Referees are to empty the
saturator completely at least once a month when three tests daily
are made, but this appears to be decidedly too long a period. The
residue of higher density gives a slightly higher candle-power
than the fresh pentane. When repeated additions of fresh fuel
are made, and the accumulated residue remains in the lamp, a
mixture is obtained such that an appreciable change in candle-
power may occur in a relatively short time. It therefore appears
desirable to empty the saturator after it has been replenished only
three or four times. It is not necessary, however, to discard the
portions emptied out ; if these are collected and not mixed with
fresh pentane, fairly reliable results can be obtained by using the
residues thus collected, making a suitable correction determined
by the density. Considerable attention has been given to the
determination of such a correction, and the results of the meas-
urements made are shown in Fig. 2. The relation between candle-
power and density is probably not really linear, but for present
purposes may be assumed so. It will be seen that the change in
candle-power averages about 1 per cent, for an increase of 0.01
in the density of the fuel. Samples of pentane from different
sources which have been initially high in density have also given
CRITTENDEN AND TAYLOR: PENTANE LAMP
427
results agreeing fairly well with this rule. Consequently it
appears to be allowable to apply such corrections in many cases
where the expense for pentane can thus be cut down materially
without any considerable sacrifice of accuracy.
In the standardization of lamps at the Bureau the pentane used
is always kept below 0.635 in density. The values certified may
therefore be considered as correct for a density of 0.630. No
correction need be made for pentane below 0.635, but above that
density correction is desirable. For example, if the pentane used
runs up to 0.650 the candle-power obtained is presumably 2 per
cent, above that certified for the lamp, and this density is likely
010.0
3.6
•
0
v^
•
^
$x
- ;
°°°
.625 .635 .645 .655 .665
DENSITY OF PENTANE AT 59° FAHR.
Fig. 2. — Effect of pentane density on candle-power of pentane lamps.
to be reached if a gallon of pentane is used without removing
the residue from the saturator.
It should be noted that the above discussion refers to pentane
which is initially fairly pure, as shown by the permanganate test.
If the fresh pentane contains impurities these are usually con-
centrated by the fractional distillation in the lamp, and the result
may be an increase in candle-power greater than that given above.
EFFECTS OF ATMOSPHERIC CONDITIONS.
It is usually assumed that the effect of all atmospheric condi-
tions is the same on all flames, and is therefore automatically
corrected when flames are compared with flames, but it is some-
times necessary to know the amount of departure from the
428 TRANSACTIONS I. E. S. — PART II
normal value, as when flame standards are compared with electric
lamps. The dependence of flame intensity on the composition of
the air is so complicated a matter that the determination of exact
corrections for vitiation of the air is impracticable, and when
observations are to be reduced to normal values good ventilation
is indispensable so that no such corrections shall be needed.
With pure air, however, there are considerable variations in
candle-power because of variations in the amount of moisture and
in the barometric pressure.
TEMPERATURE.
Temperature might also be expected to affect the pentane lamp
appreciably, but nearly all investigators have agreed that within
the usual laboratory range it does not do so. In the work at the
Bureau there has so far been available no means of changing
temperature and humidity independently, and the effects of the
two cannot be separated with certainty. The results obtained can
be represented about equally well by assuming that temperature
has no effect or by making a small correction for temperature
and using a correspondingly different factor for water vapor. In
accordance with the usual custom the former practise will be fol-
lowed, and the present discussion will be limited to the effects of
barometric pressure and of water vapor.
BAROMETRIC PRESSURE.
In general, flames give less light when the barometric pressure
is low, but the effects on different sorts of lamps are markedly
different. For the 10 candle-power pentane lamp a change of
0.6 per cent, per cm. (1.5 per cent, per inch) was found at
the German Reichsanstalt,s while the result obtained at the
English National Physical Laboratory9 is 0.8 per cent, per cm.
(2 per cent, per inch), and more recent work10 in England
has verified the latter value for pressures near the normal. Vari-
ous determinations at the Bureau have not been consistent, chiefly
because of the small range of pressure obtained. The results for
the determinations which should have been most reliable have
8 Jour, fur Gas. u. Wasser., Vol. XL,IX, p. 561. 1906.
9 Electrician (London), Vol. UJI, p. 571, 1904. Journal Institution of Electrical Engineers ,
Vol. XXXVIII, p. 271, 1906-7. Journal Gas Lighting, Vol. XCIX, p. 232, 1907. N. P. t,.
Collected Researches, Vol. Ill, p. 49, 1908.
10 Butterfield, Haldane, and Trotter, Journal Gas Lighting, Vol. CXV, p. 290, 1911, and
Ameiican Gas Light Journal, Vol. XCV, p. 145, 1911.
CRITTENDEN" AND TAYLOR: PENTANE LAMP
429
varied from 0.6 to 0.8 per cent., and consequently the English
value, 0.8 per cent, per cm. has been used.11 The chart of devia-
tions given in Fig. 3 is plotted on this basis.
WATER VAPOR DETERMINATIONS.
Water vapor in the air lowers the intensity of the flame; the
effect has been found to be proportional to the amount of water
present. This amount is expressed in liters (/) of water vapor
per cubic meter of dry air, in other words in parts of water
4
[DRY BULB TtMftRATURE6'
TT
1 i
:
w
fj
//
!
!«
/>
j
z
<
4
<
!<
z
g«
g"
[
7
If
u
i
i\
R!
-
:■:
1
/
7
h
/
L
n
1 1
H
1 f
1
i
/
1
j J
7
b s
AODIT
DUE
S 5 2 1 -0<
IOBHWETER
/
1
/
f-'ig. 3-
Variation of candle-power of pentane lamps with humidity
and barometric pressure.
vapor per 1,000 parts of dry air. The pressure (e) of the water
vapor is determined by means of a hygrometer (preferably of
the ventilated type) using tables adapted to the particular type
of hygrometer. If the barometric pressure is represented by b,
e
1 =
X 1,000.
b — e
A normal value (n) for the amount of water vapor must be
chosen more or less arbitrarily. Then if I„ is the normal inten-
sity of a given flame (that is, the intensity when there are n liters
of water vapor per cubic meter of dry air), the intensity I at any
particular time is given by the equation.
1 = 1, [1+ (;; — /) a].
11 In the Transactions of the Illuminating Engineering Society. Vol. V, p. 776, 1910,
it was inadvertently stated that the factor 0.6 is used, when it should have been said that
this factor was obtained from the data then reported on.
430 TRANSACTIONS I. E. S. — PART II
The normal proportion of water vapor has been fixed at 8.0 liters
for the pentane lamp ; the correction factor a has been several
times determined, but the agreement between results in different
laboratories is not close. These determinations were rather fully
discussed in one of the earlier articles mentioned12 ; it is sufficient
to recall here that the National Physical Laboratory13 found a to
be 0.0066, whereas the Reichsanstalt13 obtained 0.0055 and the
Bureau of Standards 0.0057. The more recent English tests13
have given 0.00625.
The method of testing lamps at the Bureau furnishes continual
data for the redetermination of the correction factor, and com-
plete calculations have been made using the results up to the
beginning of the present year on all lamps whose tests have
included a range of 5 liters of water vapor or more. To reduce
the labor of calculation the values obtained each time a lamp is
placed on the photometer have been grouped together. Some
insignificant changes in the data published in 19 10 have been
made to correct for slight errors introduced by the barograph
then used, and the revised results are given in Part I of the
table. These data include 628 groups of sets, or about 30,000
Water Vapor Correction Factors.
(I. — Observations of 1910.)
I,atnp Times en photometer Factor a Weight
Chance 1 16 74 0.00568 174
Chance 118 23 0.00565 22
Sugg 171 62 0.00565 130
American 25 . 16 0.00584 11
American 74 19 0.00564 48
American 157 11 0.00581 8
American 162 7 0.00572 9
Weighted mean value of a 0.00567
(II.— Observations of 1911-1912.)
Lamp Times on photometer Factor a Weight
Chance 116 135 0.00586 226
Chance 118 177 0.00552 339
25 other lamps 264 0.00569 269
Weighted mean value of a 0.00567
individual settings of the photometer. The data obtained from
later tests are summarized in Part II of the table ; they represent
12 Transactions Illuminating Engineering Society, Vol. V, p. 766, 1910.
18 loc. cit.
CRITTENDEN AND TAYLOR I PENTANE LAMP 43 1
about 2,300 groups of sets or perhaps 75,000 individual photom-
eter settings. In assigning weights to individual lamps allowance
was made both for the number of times on the photometer and
for the range of humidity covered. Since ranges as small as
5 liters were included, giving a total variation of only 3 per cent,
in these cases, and some of the lamps were on the photometer
only six times, the results from individual lamps cannot be ex-
pected to agree very closely. However, only two lamps gave a
value for a below 0.0050 and the highest value obtained was
0.0066. The mean result checks the former one even more closely
than could be expected, and there seems to be no room for doubt
that, at least in the Bureau laboratory, the effect is very definitely
and consistently represented by a factor of 0.57 per cent, per liter
of water vapor. This factor has been used in plotting the chart
which is reproduced in Fig. 3.
This chart has been so plotted that the departure of a lamp
from the normal value can be read directly from it when the
barometric pressure and the readings of the wet and dry bulb
thermometers of a ventilated hygrometer are known. The hygro-
meter may be either a sling psychrometer such as is used by the
U. S. Weather Bureau or a mechanically ventilated instrument
like the Assmann psychrometer. If an ordinary stationary hygro-
meter is used the covering of the wet bulb should be only one
thickness of very thin material and the reading should be taken
at the lowest point to which it can be brought by vigorous
fanning.
The curves at the left of the chart give the percentage devia-
tion from normal candle-power which corresponds to given
temperature and wet bulb depression when the barometric pres-
sure is normal, which for the pentane lamp is 760 mm. When
the pressure is different from this the additional deviation can
be found from the curves at the right as follows : First read off
the deviation in the regular way as if the barometer were normal,
and note this; then from this point pass horizontally across the
sheet to that curve at the right which represents the actual pres-
sure. Vertically above or below the point at which this curve
is reached will be found the amount to be added to the deviation
as read from the first curves. This added amount is made up
432
TRANSACTIONS I. E. S. PART II
chiefly of the direct effect of pressure on the candle-power, but
it includes also a proper allowance for the fact that the amount
of water vapor which corresponds to a given pair of bulb read-
ings depends somewhat on the barometric pressure.
The chart is plotted for the pentane lamp, but so far as we know-
it may be applied to other flames without introducing serious
errors. It would certainly be justifiable to measure gas flames.
for instance, with electric standards and to correct the observed
candle-powers to obtain normal values according to the chart.
In order to give some idea of the amount of variation in
water vapor which occurs from season to season, data calculated
from the records of the Weather Bureau which give the state of
~Z8
SE4
"16
i
fe4
£
?/
\
&
y/
\
<£
y
/
*
— -
\
y
A
4:
}y
a^;
3
^
4
&
/y
CHI
:a«
)-'
*V
A
4/
ffc
\
JP,
v'/
,m \ 2
&
/.
7?
>,
y/Z_
N
V
/,
^
X
^
5*.
E^j
JAN. FEB. MAR. APR. MAY JUNE JULY AU6. SEP OCT. NOV. DEC
Kig. 4. — Average water-vapor content of the air at various American
cities (1904 to 1908).
the out-door air have been plotted in Fig. 4. In general the
amount of moisture is small in winter and large in summer; the
average for May and for October is practically the same as for
the whole year. The variation from day to day is often great,
especially in the spring and fall, but the monthly averages show
considerable regularity from year to year. The curves are plot-
ted from the averages of observations for 5 years.
The normal values adopted in Europe are lower than the
average for nearly all parts of the United States ; consequently
the average of uncorrected candle-powers of flame standards in
this country runs somewhat lower than the normal intensities.
CRITTENDEN AND TAYLOR : PENTANE LAMP 433
The following table gives the average amounts of water vapor
at several cities for the period of 5 years plotted in Fig. 4. The
last column gives the corresponding departure of a pentane lamp
from its normal value, which is practically the percentage of
difference between the nominal and the actual average candle-
power of any source which is measured by the lamp.
Average Water-vapor and Pentane Candle-power in the
United States.
Per cent, departure of
Average water-vapor pentane lamp from
Place Liters per cm. normal value
Boston 9.9 — 1.1
Omaha 10.3 1.3
Chicago 10.6 1.5
Philadelphia 11. 2 1.8
San Francisco 11.3 1.9
Washington 12.0 2.3
New Orleans 19. 1 6.3
SUMMARY.
In this paper we have attempted to describe fully the method
of testing pentane lamps at the Bureau of Standards, to give
brief general directions for the use of the lamps, to discuss more
fully the question of fuel, and to furnish data on the effects of
pressure and moisture in a form which may be useful to those
who have occasion to reduce observations on flames to normal
candle-power values. Such new data as has been introduced has
been obtained in the laboratories of the Bureau, and our
acknowledgements are due to Dr. Rosa, under whose direction
the work has been carried on, and to Mr. G. J. Schladt, who has
assisted in most of the tests.
DISCUSSION.
Mr. C. O. Bond : We are very fortunate indeed to have
another study, or a continued study, of the pentane lamp from
the Bureau of Standards. There are three things brought out
in this paper that I would hardly have suspected, yet I have
learned that when the Bureau of Standards speaks, it is danger-
ous to hold opposite opinions on the subject unless extremely
one is well fortified, because they investigate very thoroughly.
On the ninth page it is stated that the conviction has been
434 TRANSACTIONS I. E. S. — PART II
strengthened "that most of the change in candle-power caused by-
variation in conditions of operation may be attributed to the air
circulating system and the variations in the flow of air which
arise from changes in the relative temperatures of parts of the
lamp." In the laboratory with which I am connected we have
tried to study the effect of changing the temperatures of parts of
the lamp and could not arrive at very definite formulae. We knew
that we did get variations in lamp values depending on whether
these different parts were changed in temperature. For instance,
we had one lamp that was mounted on the tripod with a solid
rod supporting it, and another one with a hollow rod supporting
it, and we got two different values, apparently depending on the
fact that one rod was a better conductor than the other one. By
abstraction of heat they would work in such a way as to change
the flow of the hot air which leads to the inner part of the flame,
and hot air determines the degree to which the pentane flame
shall bulge. In this way the air changes the area of the flame
and consequently its candle-power. Where a booth is built
around a lamp, I would have expected to see some change in the
flow of air at least as supplied to the flame exterior and that this
would have an effect in changing the candle-power of the lamp.
The second point is in regard to the change in pressure in the
saturator box. We were fixed in our belief that we did get a
considerable change, although it must be said here that the largest
changes came when we were using a regulating valve placed near
the burner. The lamps stand some thirty inches high and the
regulator cock is now placed high, at the outlet of the saturator
box. If one uses the regulator near where the vapor tube enters
the burner that is a somewhat different affair. My experience was
that if the pressure in the saturator box is increased and then
regulated by this lower valve, when the vapor passed into the
burner, it had a higher velocity, and that the effect of this velocity
had not entirely died away before the issuance of the vapor
through the perforations constituting the burner tip ; so that
there was apparently a stiffening of the sides of the flame; the
blue part of the flame went higher with the added pressure and
by thus decreasing the luminous area, the candle-power was
lessened.
The saturator box was certainly intended in the beginning, as
PENTANE LAMP 435
Dr. Harcourt has said, to be used with its inlet cock open, thus
maintaining a constant pressure head. I have always gone on
the assumption that with a constant pressure head, with uniform
fuel fed through a constant orifice, there would be a constant
consumption, and therefore a constant flame, leading to a con-
stant candle-power.
The third point which is of great interest and which is very
important indeed is the quickness with which they claim pentane
as now furnished to the market seems to change its density.
Mr. Crittenden has said that after one-tenth of it has been used
the residue is already well beyond the limits set for the
density of the pentane. It hardly seems possible that this could
have been true at the time Dr. Harcourt invented his lamp and
made the specification for its use. I am wondering whether it
is not now possible to meet the tests provided in the specifica-
tions and yet not supply the exact article that was intended in
the original specifications. If I remember aright, the specifica-
tion of light American petroleum such as is known as gasoline
from which to refine the pentane, referred originally to Pennsyl-
vania oil with a paraffin base. This is not so readily supplied
now as formerly, and it may be due to this lack of a paraffin
base that the Bureau has found such a rapid change in the pen-
tane density. I do not claim any knowledge in this direction
(but the matter is of such importance to the reputation of the
pentane lamp as a standard of reference, that the subject should
be investigated further, and access be had to original records in
England to find where this discrepancy has crept in).
Mr. C. W. Jordan : I would like to inquire in regards to the
relative accuracy of carbon dioxid determinations in air with a
Zeiss refractometer and by chemical methods. It occurred to
me that the refractometer shows its greatest accuracy when only
two constituents of a gaseous mixture are variable. By Petters-
son's chemical method, carbon dioxid in air can be determined
accurately to within 0.002 per cent.
Have the authors of this paper made' any extensive investiga-
tions of pentane from a chemical standpoint? While the fuel
consumed in the lamp is designated by the formula Cr,H12 it is
admittedly not pure pentane. In fact the portions distill off
436 TRANSACTIONS I. E. S. PART II
between 25 ° C. and 400 C. which precludes a definite compound.
Prepared from the rather vaguely defined American petroleum,
it is said to consist chiefly of pentane with small quantities of
lower and higher homologues whose presence does not affect,
seriously, the candle-power of the lamp. Pentane is the fifth of
the paraffin series of hydrocarbons, the first four of which,
methane, ethane, propane and butane being gaseous at ordinary
temperatures. Methane burns with a non-luminous flame and
the candle-power of the succeeding compounds increase pro-
gressively, due to the increase in the percentage of free carbon
dissociated. Pentane is known to exist in three modifications,
each having different boiling points, and the separation of admix-
tures of propane, butane and the higher homologue, hexane, is
very difficult.
For lack of evidence to the contrary, I am inclined to believe
that serious variations in candle-power may occur in different
lots of pentane having the proper specific gravity and giving
negative tests for the hydrocarbons of the benzene series. I
think that flame standards should consume a definite, chemically
pure fuel and in the case of the Harcourt pentane lamp, syn-
thetically produced pentane would be the ideal fuel.
On the fourth page of this paper a statement is made that pen-
tane lamps, manufactured by the same maker and supposedly of
the same type, sometimes show variations in candle-power of as
great as 4 per cent. This variation is believed to be due to
mechanical differences in the lamps. While it is exceedingly
important to standardize lamps under the existing conditions, I
think that the line of future investigation should be that of draw-
ing up standard specifications to eliminate these differences.
Mr. E. C. Crittenden (in reply) : Mr. Bond's complimentary
remarks regarding the work done on pentane lamps at the Bureau
of Standards are very gratifying. It is a pleasure also to
acknowledge our indebtedness to Mr. Bond for valuable sugges-
tions which he has given us at various times during the progress
of this work. The position which the pentane lamp now holds
in this country is due in no small degree to the pioneer work
which he did in introducing it, and his thorough personal knowl-
edge of the lamp gives much weight to his suggestions.
PENTANE LAMP 437
It has been stated that the relative temperatures of various
parts of the air circulating system of the lamp are probably the
cause of most of the changes in candle-power which occur when
the lamp is operated under different conditions. The case in
which a change of candle-power was brought about by substi-
tuting a solid rod for the usual hollow one at the base of the
lamp is an interesting example, which would be hard to explain
in any other way. Since the temperature may be appreciably
affected by apparently unimportant changes in the conditions of
operation, it is evident that if the lamp is to be enclosed, great
care must be taken to make sure that the enclosure does not
affect the candle-power. As is stated in the paper, it is better not
to use any screening except that necessary to shut out stray light.
The additional screens described were used because of unfavor-
able conditions in the laboratory, which would otherwise have
prevented accurate measurements at times when drafts were bad ;
the particular screen described was found by comparative meas-
urements to give no perceptible effect on the candle-power.
With regard to the change in density of the pentane and the
possibility that the pentane commercially supplied at present is
not the same as when the standard was adopted by the Gas
Referees, because the source of supply is different, I would say
that if the material is prepared according to the Referees' direc-
tions the products obtained from various kinds of crude oil ought
not to differ materially. Chemists who have studied various oils
say that the differences between them are largely in the heavier
constituents ; so far as the very light fractions are concerned,
Western oils are not essentially different from Pennsylvania oil.
In the paraffin series the next substance above pentane is hexane,
and that form of it which appears in petroleum has a boiling
point of about 6o° C. The four distillations specified below that
temperature ought to remove the hexane quite completely. The
final distillation is carried over the rather wide range from 25 °
to 400 because the pentane itself exists in two forms having
boiling points about 280 and 360 C. Incidentally, the densities
of the two forms are 0.625 and 0.631, and it is not clear why the
referees set the narrow limits of 0.6235 to 0.626 in the specifi-
cations.
There is little danger of errors arising from failure to remove
5
43§ TRANSACTIONS I. E. S. — PART II
the other members of the paraffin series, but amylene (C5H10),
of the define series, cannot be separated from the pentane by
distillation alone, because it has several forms whose boiling
points come between 25 ° and 400 C. The sulphuric acid treat-
ment prescribed should, however, remove it; the permanganate
test is supposed to show whether this has been properly done.
It is by no means certain that the pentane on the market is
prepared in accordance with the Referees' directions. To make
up a mixture which shall have the required density is not diffi-
cult, and consequently it is desirable that the chemical test be
made.
By repeated distillation, or by synthesis, it is possible to obtain
practically pure pentane, but the cost would prohibit its use in
testing. If it were a matter of setting up a primary standard
the pure fuel might be prepared. There are, however, other
difficulties to be met before the lamp can be seriously considered
as a primary standard; some time it may be so perfectly under-
stood that we shall be willing to base the unit upon it, but per-
sonally I would say that our hopes of attaining that end are not
high.
T. J. Little, Jr. : I notice on the twenty-third page it is stated
that "The chart is plotted for the pentane lamp, but so far as we
know it may be applied to other forms without introducing serious
errors." I suppose that is meant for the naked flames, as for in-
stance, flat flame gas burners, but in the accurate determination
for candle-power of incandescent mantle burners such as those
for use on either gas or gasoline, would you also recommend cer-
tain correcting: factors?
KINGSBURY: SUNDAY-SCHOOL ROOM ILLUMINATION 439
EXPERIMENTS IN THE ILLUMINATION OF A
SUNDAY-SCHOOL ROOM WITH GAS.*
BY EDWIN F. KINGSBURY.
Synopsis: The author describes the lighting by gas of a typical
Sunday-school room, composed of a large central floor with a high glass
paneled ceiling and alcoves at the ends. Four 4-burner "arcs" having a
magnet-valve pilot system of control from the main floor were placed
above the ceiling to light the central area. The alcoves were lighted from
the side walls by placing a large semi-circular translucent screen in
front of each of the eleven upright burners. The object of these screens
was to transmit part of the light directly and to reflect a portion back
on to the wall. The whole forms a light source of large area and low
intrinsic brilliancy.
When planning the lighting of a church, the auditorium is
usually carefully studied from every angle to secure ample, uni-
form illumination and to bring out the best artistic effect. By
the time the Sunday-school room is reached ideas of economy
become strong and almost any style of illumination is made to do,
though this room may actually be used more frequently and for
more varied purposes than the auditorium. While, from the
nature of the two rooms, it may not be desirable to aim for the
same effect in the Sunday-school room as in the church proper,
still the lighting of the former should be carefully studied to in-
clude the best principles of a good practical installation.
The purpose of this paper is to describe some experiments in
the illumination of the Sunday-school room of the Summit Pres-
byterian Church, of Germantown, Philadelphia.
This room is typical of many used for Sunday-school pur-
poses, being composed of a large, open central portion with
alcoves at two sides which can be closed and utilized for classes.
The lighting requirements for such a place demand that the
sides shall be treated more as distinctive rooms and that two
separate plans of lighting be adopted. The most usual solution
*A paner read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
440 TRANSACTIONS I. E. S. — PART II
provides one or more chandeliers over the main floor and a
smaller one in the center of each alcove. In the present case
there was a stained glass ceiling over the center which could b'e
used advantageously and in each alcove there were two wall out-
lets that it was desirable to use.
There were several reasons for placing the light sources above
the ceiling in the central portion. First, it removed the light from
the field of vision and, secondly, it utilized the beauty of the
stained glass, which any lamps hanging below would tend to ob-
scure. Then, also, the attic was easy of access for maintenance
purposes.
The problem of lighting the sides was a more difficult one to
solve, as, utilizing the present outlets, brought the lights too low,
especially as an audience sits where the light from one row of
lamps would be square in the eyes. Even well frosted globes were
objectionable, as the walls were dark and the contrast strong.
In accordance with what has been said the work may conven-
iently be divided into two portions — the first dealing with the
permanent lighting of the central portion of the room through
the stained glass ceiling and the second with a temporary instal-
lation along the front and rear to illuminate the alcoves.
The plan of the room, with essential dimensions, is shown in
Fig. i. The room is 60 ft. (18.29 m-) l°ng by 35 ft. (10.67 m-)
wide and 35 ft. (10.67 m-) high. Twelve feet (3.66 m.) from the
front and back walls open archways rise, the walls in front at
the top curving inwardly to meet a glass paneled ceiling 28 ft.
(8.53 m.) long by 42 ft. (12.80 m.) wide, each panel being 3 ft.
by 3 ft. (0.91 m. x 0.91 m.). Heavy blinds slide between the arch
posts at the rear to form three separate rooms. The wall north
of the main floor, composed of three heavy Venetian blinds which
can be raised to include the low-ceiling room beyond it as a part
of the school room, rises to only one-third of the full height of
the room, the space above being entirely open. These three blinds,
however, were closed throughout the test. At the south are large
double sliding doors in the center, the body of the wall being
plain and curving inwardly at the top to meet the ceiling, as in the
front and rear. Wooden wainscoating 5 ft. (1.52 m.) high ex-
KINGSBURY: SUNDAY-SCHOOL ROOM ILLUMINATION 44I
tends around the room, the north side excepted. All wood work
is dark oak in color and the walls dark green.
The room is lighted in the daytime by three large stained glass
windows at the rear. The three windows in front are now of
no use, being covered by a later built portion of the church.
The body of the room was originally lighted by 12 small in-
verted incandescent gas burners at each end, distributed uni-
formly along each horizontal beam 15 ft. (4.57 m.) high at the
base of the arches and supplied with pilots and pendant chains for
lighting.
SLIDING DOORS I.. '.'. ' 1
Fig i.— Plan of room.
The illumination of the alcoves was provided by three small
inverted units and several open flame burners. This installation
was unsatisfactory, for at least two reasons. First, the mechani-
cal operation of reaching up with a hook to catch 24 rings was
both monotonous and time-consuming, especially if the room was
rather dark at the start, or if many rings were accidentally set
to swinging before being caught. Secondly, the glare from the
row at one end in the eyes of a person sitting at the other end was
annoying.
These two rows of units lighting the body of the room were,
442 TRANSACTIONS I. E. S. — PART II
therefore, removed and the glass paneled ceiling utilized by plac-
ing a four-burner inverted incandescent gas arc lamp above each
of four panels, as indicated by the circles in Fig. i. The magnet-
valve pilot system of ignition is used. Two push buttons are lo-
cated on the main floor near the southeast door ; each one lights a
pair of lamps. Each lamp is equipped with a polished aluminum
conical focusing reflector n in. (27.94 cm.) diameter at the top,
26 in. (66 cm.) at the bottom and 16 in. (40.64 cm.) high. The
mantles are 30 in. (76.2 cm.) from the glass ceiling and the bot-
tom of the reflector 16 in. (40.64 cm.). Opal globes were decided
upon as the most suitable. Clear globes gave a bright spot in the
center with bright concentric rings. The spots were not visible
without the effort to look up at a high angle at the lights, but the
illumination on the floor below any lamp was nearly double that
half-way between. Opal globes, however, obviated this difficulty,
as shown in Figs. 4 and 5. This system has been in use a year
and a half and has given satisfaction.
The desultory lighting in the rear of the arches was untouched
until recently, when the experiment described below was tried.
This consisted of placing upright incandescent gas burners on the
six outlets in the front and on five of the six in the rear. Ten
inches (25.4 cm.) in front of each lamp was a semi-circular trans-
lucent paper screen, highly glazed on the inner side and matt on
the outer. A plain creamy-white bathroom wall paper answered
the temporary purpose very well. The object of this was to have
the glazed surface reflect part of the light to the wall. Then the
light reflected from the wall, with that fraction transmitted by the
screen, is supposed to form a continuous light source of large
area and low intrinsic brilliancy. The position of the curve is
imitative of daylight window illumination and should partake
of the naturalness of the latter. If the wall is ornamental the
screens may well be. In the present case both were left plain.
The screen, ten inches (25.4 cm.) high with a radius of ten
inches (25.4 cm.) and having the light at the center, shields the
source from practically every position a person would be likely
to assume. The exception is that when one sits sideways and
almost directly beneath a lamp the bare mantle can be seen by
KINGSBURY! SUNDAY-SCHOOL ROOM ILLUMINATION 443
looking upward. This can easily be remedied by making the
screen a little wider or lowering the present one somewhat.
In order to prevent too much light being thrown back against
the wall and to direct more obliquely to the floor, as sunlight
would be, a small rectangular piece of bright tin was placed
(experimentally) directly in the rear of each mantle, allowing it
to slant forward a trifle. By selecting the proper width and angle
for such a reflector any desired effect in this connection could be
obtained.
Owing to the high absorption of the wall and woodwork, the
employment of a more efficient diffusive surface back of each unit
was imperative to secure sufficient illumination. For this pur-
pose a semi-matt white paper 36 in. by 36 in. (0.913 x 0.913 m.)
was pasted on the wall.
Fig. 2 shows the front of the room and Fig. 3 the rear under
this artificial illumination. In the former can be seen the light
reflected from the floor, which appears as a bright band running
parallel with the wall. It will be noted at the top of the picture,
above the horizontal beam, that there is considerable light thrown
obliquely upwards and a design painted on the under side of the
arches was brought out conspicuously clear. Unfortunately the
wall near the ceiling could not be changed, but if it were white or
a light cream it would help out in a line where most needed —
the transition from the side to the main overhead illumination.
The uniformity of the screens and background is shown in
both pictures where they cannot be distinguished, except at an
oblique angle, as in Fig. 3. This system seems to solve success-
fully the illumination of the alcoves.
In Fig. 6 is shown a candle-power distribution curve of one
of the wall units, taken through the plane bisecting the central
angle of 1800 subtended by the screen. Each arm of the
curve has its own particular function. The two back loops go to
the wall to be reflected, one to the floor and the other at an
oblique angle toward the ceiling, while a considerable part of
both will be diffused into the space in front of the unit. The
upper front one goes to the ceiling and the lower front one to
the floor. It is desirable to have the curve directly in the rear of
444 TRANSACTIONS I. E. S. PART II
the screen swing in more, as the light thrown straight back is
more or less useless. This can be partially accomplished by mak-
ing the tin reflector larger, or by having the semi-circular screen
not as a section cut from a cylinder, but as a section of an hour-
glass. Even then any diffused light would keep the wall quite
bright.
Illumination measurements were taken with all the room fully
lighted, and with the center lamps and the lamps at one end
lighted separately, the object being to show the effect of the two
distinct systems on each other. All readings were taken in a 30
in. (76.2 cm.) plane, with the exception of the readings on the
platform, which were made on a plane 2 ft. (60.96 cm.) higher
than those on the floor.
Fig. 4 shows the first case when all the lamps are at maximum
brilliancy. This seems to give the best effect and with the light
properly spread no fatigue or annoyance should be experienced
by an audience facing them for several hours. However, the
alcove lighting is very elastic, and if the maximum light desired
at one time is considered excessive at another time, the lamps can
easily be turned down, as they are upright burners and will work
equally well at any intensity.
An analysis of the results shows a high candle-power close to
the lamps, with a rapid falling off to a minimum directly beneath
the arches. Then comes a gradual rise to a maximum on the
main floor.
Fig. 5 shows the center and one end lighted separately. As
will be noted the end wall illumination has an appreciable effect
on the main floor, the result being to enlarge the contours. It
would be more desirable if the ends sent more light beneath the
arches, and this could easily be secured by making the entire
wall much lighter.
The average foot-candles on the 30 in. (76.2 cm.) plane in the
center is 1.08. While this may seem low, it has proven to be
ample for all purposes and there seems to be a peculiar advantage
in reading responses or singing where the book is held fairly hori-
zontal. The vertically directed light here from a height gives one
a feeling the same as that experienced in cathedrals and is hence
l^
Fig. 2. — Front of room lighted.
y
L k
•f !
n
^1
^'"i*"*— r;
—
Is -
r
^''oWi*
1 *
hliMfl
■.■■-«^.
Fig. 3. — Rear of room lighted.
«0 J 'tlS
~~i7~Jrri; iar^~n
fc-
/^?0 OS 7
^V7\ MOB:
- » f - S^ODE
\ ' A Nl >■'»-,— ->J- . <
>X \ \ ( \ -<^„
■/> -)l-9i zis^ ' '061 \ aw v ~" ~ai "" "
^b W777ZW///////////////A
x i^DDidQft]
"DTsoiio
KliQJDiDlnTaa
E^
1 W
\ s'975 W0\ v 400
W ^
tit
PLANE OF READIN6S 30' HIGH
Fig. 4. — Illumination readings.
ALL LIGHTS OK FULL
.~)-1.80 I 1.60
! [oh
4.00 \ [ahe a30
□HHBQi
oBy'SB
01
0
^/HTToo £I5> J>
isn
so
Op
10000
-#-
N
•~}2.90 ' 1.70 v' C.50
-~ J.jJ0_
0.85 0.95
<1
■
I
'i ! ,
PLANE OF READING 30"HI6H. END &CENTER LIGHTS ON SEPARATELY AS MARKED.
Fig. 5. — Illumination readings.
KINGSBURY: SUNDAY-SCHOOL ROOM ILLUMINATION
445
appropriate. Experience has shown the body of the room is suf-
ficiently well lighted by the four arc lamps, but they have
little influence back of the arches.
There are spots under the archways where the illumination is
a trifle too low to allow of continued reading, but these would be
corrected by lighter side walls in a finished installation.
This installation is described with the purpose of showing that
in rooms so cut up and apparently hard to illuminate well, the
very peculiarities may be used to provide something at once un-
usual and practical. Certainly, the illuminating engineer should
not feel that any situation is so commonplace and unimportant
r /\/ y\ ^-V-"
— *"f»v 7\ \/\ s
80 I 60 [ 40 | V fl
y^ZO |40-4v60 8.0
XXfRont^^/^
_^J\r-^^ACK/<C/ ,
Fig. 6. — Candle-power distribution through vertical
plane of an alcove unit.
that any casual style is good enough, but he should be alive to
every opportunity to make the uncommon out of the common.
This means that while there are certain fundamental rules always
to be observed, they make only a beginning, and the remainder
of his task depends on his originality. This is not assuming, of
course, that expense is to be overlooked. In the case described
above a relatively costly installation was a poor one. It means
that thought and study must be given to each individual case.
The thanks of the author are due to Mr. Charles O. Bond, Dr,
Herbert E. Ives and Mr. C. W. Jordan for assistance in prepar-
ing this paper.
44^ TRANSACTIONS I. E. S. — PART II
DISCUSSION.
Mr. J. D. Israel: I feel privileged to discuss the question
of church lighting from a commercial standpoint. The psycho-
logical effect of lighting, discussed by Dr. Lewis* yesterday,
I think should be considered when one studies church lighting
problems from a purely commercial standpoint.
I wish to say for your information that the Philadelphia
Electric Company approaches this problem purely on a com-
mercial basis, the same as they go after business in any other
class of lighting. We address letters to the trustees and pas-
tors of the various churches; we circularize and give illustra-
tions of such installations as we already have connected to our
circuits. Then we follow with personal appeals; our men
appear before the trustees. We ask the pastor and trustees to
consider the matter as a commercial problem, as an aesthetic
problem and as a religious problem. We accomplish good
results without detracting at all from the sacredness of the
work.
Mr. E. F. Kingsbury: One use of the alcove lighting not
mentioned in the paper is in the illumination of small rooms,
especially in residences, where there are wall outlets that one
dislikes to use on account of the difficulty in avoiding objection-
able glare from them. With a little care the light thrown
downward can be utilized for reading on a table near the wall
and the light thrown upward will supply the general illumina-
tion. The success of such an installation depends on keeping
the brilliancy of the screen and reflecting walls low. This is
well secured on the screen and an artistic end realized if it is
made ornamental. In this way an almost useless outlet might
be turned into a thing of beauty.
* "The Psychic Values of Light, Shade, Form and Color," Trans. I. E. S., p. 357
(October, 1913).
LANSINGH : ENCLOSING GLASSWARE 447
CHARACTERISTICS OF ENCLOSING GLASSWARE.*
BY VAN RENSSELAER LANSINGH.
Synopsis: In this paper, enclosing glassware is divided into two
classes : one, purely transmitting and diffusing, such as ground, opal and
leaded glass; the other, prismatic glass, which employs the principle of
specular reflection. Photometric curves and data are given to show (r)
that with glassware of the first class, little except good diffusion and
low absorption may be expected, the re-distribution of light being negli-
gible; (2) with prismatic glass, the distribution may be varied in accord-
ance with the wishes of the engineer. The absorption in both given classes
is about the same.
The two most pronounced tendencies of the day in interior
illumination, excluding the industrial field, are, first, the use of
indirect and semi-indirect lighting, and second, enclosing glass-
ware. By the latter is meant the use of large diffusing glass-
ware, completely surrounding the lamp. A great deal has been
written in the technical press and elsewhere on the first class,
but little has appeared regarding the second. The present
paper, therefore, aims to state briefly some of the chief character-
istics of this class of lighting units. So far the writer has made
a study only with tungsten-filament lamps, but with the possible
exception of some changes in color characteristics, it is believed
the same results would be obtained with gas mantle burners.
This cannot be said of arc lamps, however, when the position of
the arc travels, and no attempt is made here to cover this phase
of the subject.
The chief characteristics of enclosing glassware are: (1) dis-
tribution, (2) absorption, (3) appearance, (4) effect on the eye,
(5) effect on the lamp, and (6) color.
The question of appearance, color, and effect on the eye are
not taken up here, as these subjects have been covered by
numerous writers in the Transactions and elsewhere. The effect
on the lamp need not be given much consideration as a number
of tests have shown that the rise in temperature, due to enclosing
the lamp, is not sufficient to affect its' life.
The tests reported in this paper were made at three different
* A paper read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
448 TRANSACTIONS I. E. S. — PART II
laboratories, whose methods and results have been carefully
checked against each other ; so that the curves given are all com-
parable. Inasmuch as comparative rather than absolute values
are desired, the curves are given without the actual candle-
power. All curves, however, are plotted on the basis of 1,000
lumens for the bare lamp, i. e., the candle-power actually found
at every reading in the tests is multiplied by the ratio of 1,000
to the actual lumens of the lamp. Thus, in the case of a 100-
watt lamp giving 908 lumens, the candle-power readings would
be multiplied by ■ ' =1.1. By thus reducing all the curves
to the basis of a 1,000 lumen lamp, the curves can be compared
without reference to the size of lamps used. For example, it
will be noted that the bare lamp curve is the same in every case,
irrespective of the size of lamp tested. In every case, the fig-
ures given are based on lamp flux rather than emitted flux, as the
engineer is really concerned with the actual flux of a lamp which
is available for any given purpose, rather than the relative
emitted flux in different zones.
There are many different kinds of enclosing glassware on the
market at the present time, and typical examples from the dif-
ferent classes were selected for the purpose of the above men-
tioned tests. The kinds of glassware selected were as follows :
Class I — Pressed opal ball in two pieces.
Class II — Blown opal ball.
Class III — Blown opal acorn.
Class IV — Cased opal ball.
Class V — Leaded opal ball.
Class VI — Ground glass ball.
Class VII — Prismatic deep reflector-bowls.
Class VIII — Prismatic shallow reflector-bowls.
Class IX — Prismatic reflector-balls.
The actual trade names, designations, etc., of the glassware
tested, and full data on each test, are given in an appendix to
this paper, for those who wish this information. The accom-
panying illustrations will serve to identify the general appearance,
contour, etc., of each unit.
LANSINGH : ENCLOSING GLASSWARE) 449
Fig. I shows a two-piece pressed opal ball of comparatively
light density, listed above as Class I. Fig. 2 shows the curve
of the bare lamp and a 10-inch (25.4 cm.) ball tested with a
100-watt tungsten filament lamp. Fig. 3 shows the distribution
from a 14-inch (35.56 cm.) ball with a 100- watt tungsten fila-
ment lamp, and Fig. 4 the same with a 250-watt lamp.
It will be noted from a study of Figs. 2, 3 and 4 that a change
in the size of lamp or of the size of ball has but little effect in
the resulting distribution curves. It will be further noted that
the curves all tend toward a circular distribution, denoting good
diffusion, but, at the same time, very little redirection of light
in useful zones.
Fig. 5 is a picture of a blown one-piece opal ball (Class II),
the density being the same, but the thickness being less than in
the pressed ball just considered. Blown opal balls of this den-
sity and thickness are regularly furnished with the outside sand-
blasted or roughened to increase diffusion. Fig. 6 shows the
photometric curve of the 12-inch (30.48 cm.) size tested with a
100-watt lamp. It will be noted that the curve is less modified
from the bare lamp curve than in the former case showing that
the diffusion is not as good.
This is further emphasized by Fig. 7, the right-hand curve
being that of the pressed ball and the left-hand one the blown
ball. The difference, while noticeable, is not striking. The
use of such balls instead of the pressed type, would mean that
a greater flux would strike the side walls and less fall upon the
ceiling. An appreciably greater flux is also shown in the lower
hemisphere. This type of ball is used largely for street lighting
purposes and it will be seen that it has a low absorption, namely,
about 14 per cent, as compared with an absorption of about 24
per cent, in the case of the pressed type. A comparison of the
flux in the different zones as given under the curves shows an
increase in the flux from zero to 6o° of approximately 22 per
cent., and from zero to 900 of 19 per cent. It would seem,
therefore, that in most cases, except perhaps from the standpoint
of appearance, the blown ball, being much thinner, is preferable
for use when there is no objection to a slight image of the fila-
ment through the glass.
450 TRANSACTIONS I. $. S.— PART II
Fig. 8 shows a blown opal acorn shape (Class III) enclosing
unit made of the same glass as the others so far considered, and
roughed outside. From its distribution curve, Fig. 9, it will
be noted that due to its shape, there is a slight reflecting power
which throws more flux in the lower hemisphere. The absorp-
tion is somewhat greater than the blown ball, but less than the
pressed type, as might be expected. Compared with the blown
ball, the acorn gives about the same flux below the horizontal
but 10 per cent, more in the 0-600 zone.
Fig. 10 is a cased one-piece opal ball (Class IV) and Fig. 11
the curve of the 12-inch (30.48 cm.) size tested with a 150-watt
lamp. Comparison should be made between these and the blown
opal ball, a curve of which is shown in Fig. 6. It will be noted
that the flux below the horizontal is practically the same but
that there is less above the horizontal, resulting in an increase in
absorption of about 5 per cent. The diffusion, however, with the
cased ball is much better than in the blown opal.
Fig. 12 is a leaded opal ball (Class V), the curve of which,
with a 150-watt lamp, is shown in Fig. 13. It will be noted that
the absorption is practically the same as that of the 14-inch
(35.56 cm.) pressed ball (Fig. 3 and 4). The distribution
curve is somewhat different, however, there being slightly more
light on the horizontal and less light directly above and below.
The results of this test are more or less surprising inasmuch as
there is a general idea that most diffusing leaded glass absorbs a
large percentage of light. It would seem that this particular type
of leaded glass has no greater absorption than the ordinary light
density pressed opal ball.
Fig. 14 is the curve of a 12-inch (30.48 cm.) ground glass ball
(Class VI) with 100- watt lamp. But little alteration in the
curve of the bare lamp is made, although some diffusion is ob-
tained. The diffusion, however, is quite different from that ob-
tained with the opal ball, Fig. 6. A comparison of the two
curves shows a somewhat lower absorption in the case of the
ground glass ball than with the opal blown ball, but a somewhat
lower efficiency in the zero to 6o° zone and more light at angles
near the horizontal.
V
V*
>A
Fig. i. —Two-piece pressed opal
ball (Class I).
1
/
/
1
\ \
\ \
1 1
1
J /
Fig. 2.— Photometric curve of io-inch two-
piece pressed opal ball tested with ioo-
watt lamp.
Kig. 3.— Photometric curve of 14-inch two-
piece pressed opal ball tested with
100- watt lamp.
/
1
1
-
-
-
-
-
>
\
\
\
\
' 1 '
\
\
\
\
-
-
-
-
'
<
/
/
/
1
V / '
\ / /
Fig. 4. — Photometric curve of 14-inch
two-piece pressed opal ball tested
with 250-watt lamp.
Zonai, Flux with 1,000 Lumen Lamp.
Zone 0-60 0-90 90-180 0-180
Lamp alone 203 512 4S8 i.oco
Fig. 2 182 379 385 764
Fig. 3 190 384 335 7 '9
Fig. 4 l82 374 346 720
Fig. 5.— One-piece blown opal ball
(Class II).
x
/
/
/
\
\
\
/
/
1
I
\
l\
/
/
\
^~^-
s^^
Fig. 6. — Curve of 12-inch one-piece blown opal
ball tested with 100-watt lamp.
/ "-SBL
Fig. 8. — Blown opal acorn (Class III).
*' ^-^~T^
/^^\
X
/
\ 1
1 /
\ 1 ■
\
1 1
1 1
/ /
Fig. 9.— Curve of 12-inch blown opal acorn
tested with 150-watt lamp.
Fig. 10.— Cased opal ball (Class IV).
Fig. 11. — Curve of 12-inch cased opal ball
tested with 150-watt lamp.
D*
^o
Fig. 12.— Leaded opal ball (Class V).
Fig. 13.— Curve of 12-inch leaded opal ball
tested with 150-watt lamp.
Fig. 15.— Prismatic deep reflector-bowl
(Class VII).
/ \
! r^
y~\ \
\ /
\ / ,
\ /
V / /
\ >v /
\ / /
Fig. 16. — Curve of 14-inch prismatic deep re-
flector-bowl tested with 250-watt lamp.
Zonal Flux with 1,000 Lumen Lamp.
Zone 0-60 0-90 90-180 0-180
Lamp alone 203 512 488 1,000
Fig- 6 223 469 395 864
Fig- 9 243 476 347 822
Fig- 11 222 456 359 815
Fig. 13 181 396 321 717
Fig. 14 206 484 419 903
Fig. 16 412 589 180 769
Fig. 7.— Comparison curves of Class II Fig. 17.— Comparison curves of Class VII
and Class I, Figures 6 and 2. and Class I, Figures 16 and 3.
Zonal Flux with 1,000 Lumen Lamp.
Zone 0-60
Fig. 7— Class 1, Fig. 2 182
Class 2, Fig. 6 223
Fig. 17— Class 1, Fig. 3
Class 7, Fig. 16 412
190
0-90
379
469
384
589
90-180
3S5
395
335
180
0-180
764
864
719
769
Fig 14— Curve of 12-inch ground glass ball (Class VI) tested with 100-watt lamp.
\f
/J
Fig. 18. — Prismatic deep refleetor-bowl Fig. 19. — Curve of 14-inch prismatic deep
with special deep bowl. reflector-bowl with special deep bowl
tested with 250-watt lamp.
Fig. 20. — Prismatic shallow reflector-bowl Fig. 21. — Curve of 12 inch prismatic shal-
(ClassVIII). low reflector-bowl tested with 100-
watt lamp.
ZoNAt Fivux with 1,000 Lumen Lamp.
Zone 0-60 0-90 90-180 0-180
Lamp alone 203 5 1 2 488 1 ,000
Fig. 19 420 5S3 185 768
Fig- 21 362 554 169 723
Fig. 22. — Prismatic reflector-ball
(Class IX).
Fig. 23. — Curve of 12-inch prismatic reflector-
ball tested with 150- watt lamp.
Fig. 24.— Comparison curves of
Class IX and Class II, Figures
23 and 6.
/ \
1 /
1 \
u \ 1
1 /
\ /
\ / /
\ / /
x / /
\ 1
v v
Fig. 25. — Curve of 12-inch prismatic reflector-
ball, reflector satin finished inside, tested
with 150-walt lamp.
Zonal Flux with 1,000 Lumen Lamp.
Zone 0-60 0-90 90-1 So 0-1S0
Larnpalone 203 512 488 1,000
Fig. 23 (Class IX) •■•• 4"55 618 103 721
Fig. 6 (Class II) 223 469 395 864
Fig. 25 356 552 1S2 735
LANSINGH : ENCLOSING GLASSWARE)
451
Fig. 26.— Curve of 14-inch prismatic reflector-ball tested with 400-watt lamp
with center of filament at center of ball.
^ V
Fig. 27.— Curve of same unit as Fig. 26 but with filament 2 inches higher.
Zonal Flux with 1,000 Lumen Lamp.
Zone 0-60 0-90 90-180 0-180
Lamp alone 203 512 488 1,000
Fig. 26 239 522 '.63 786
Fig. 27 332 551 168 719
With prismatic enclosing units, a most decided change in the
resulting distribution may be noted. With opal and ground
glass diffusing units, the tendency of all curves is to become cir-
cular in shape, with increasing diffusion, but in practically no
case is there a large increase in useful flux below the horizontal.
In only three cases, Figs. 6, 9, and II, is the 6o° flux
greater than that of the bare lamp, being respectively 223 lumens,
452 TRANSACTIONS I. E. S. PART II
243 and 222, as compared with 203 for the lamp alone. The
lower hemispherical flux in every case, however, is considerably
less than that of the bare lamp, averaging 18 per cent, less, ex-
clusive of the ground glass ball.
The most pronounced redirecting effect is with the blown acorn
(Fig. 9) where 24.3 per cent, of the lamp flux is within the
O-600 zone. A comparison of this with Fig. 23 shows the re-
markable difference which can be obtained by the use of specu-
larly reflecting media. In the latter case, no less than 45.5 per
cent, of the flux is within this zone, an increase of 77 per cent.
Fig. 15 is a prismatic deep reflector-bowl, with a diffusing
bottom, a photometric curve of which is shown in Fig. 16. This
distribution is an intensive type and by far the greater flux is
below the horizontal : 41 per cent, is within the 6o° zone, and
59 per cent, is below the horizontal. The total absorption is about
23 per cent.
Fig. 17, a comparison of Figs. 4 and 16, is interesting. The
former curve, that of the two-piece pressed opal sphere, shows
only 18.2 per cent, of the flux in the 6o° zone as against 41.2 per
cent, for the prismatic deep reflector-bowl ; that is, the latter unit
utilizes about two and a quarter times as much flux in this zone.
The lower hemispherical flux is 58.9 and 37.4 for the reflector-
bowl and opal ball, respectively, with an absorption of 21.3 per
cent, and 28 per cent, in each case.
Fig. 18 shows a modification of the unit shown in Fig. 15, the
shallow bottom bowl being made much deeper in the form of an
acorn. The resulting photometric curve, Fig. 19, is not greatly
altered, although the general shape looks more like the extensive
type of distribution than the intensive type. The principal dif-
ference is the flux between zero and 300.
Fig. 20 shows a shallow prismatic reflector-bowl unit, the bowl
having shallow external redirecting prisms, and Fig. 21 the re-
sulting distribution curve. It will be noted that this gives a
broad distribution, much broader than that shown in Fig. 16 and
19, but the zonal and overall efficiency is somewhat lower. Of
the total lamp flux, 36 per cent, is within the 6o° zone
and 55 per cent, below the horizontal while the absorption is
about 28 per cent.
LANSINGH I ENCLOSING GLASSWARE 453
Fig. 22 shows a prismatic reflector-ball, the reflector being
separate from the blown globe and resting on a shoulder made
for that purpose. The lower half of the ball is satin finished,
but the part under the reflector is clear. Fig. 23 shows the dis-
tribution curve obtained from this unit with the lamp in the stand-
ard position. It shows a remarkable control of light, 45.5 per
cent, of the lamp flux being within the zero to 6o° zone and 61.8
per cent, below the horizontal ; while the total absorption is about
28 per cent. A comparison (Fig. 24) of this unit with the blown
opal ball (Class II, curve shown in Fig. 6) is even more striking
than the comparison in Fig. 17. The flux in the zero to 6o°
zone is 45.5 per cent, of total lamp flux in the case of the re-
flector-ball and only 22.3 per cent, with the opal ball, the corres-
ponding figure for the lower hemisphere being 61.8 per cent,
and 46.9 per cent, while the absolute absorption is 27.9 per cent,
and 13.6 per cent, respectively. The relatively low upper hem-
ispherical flux shown in this reflector-ball is quite remarkable
and strikingly illustrates the light control which is possible in
prismatic combinations, even with a secondary globe between the
light source and the active reflector as is the case in the larger
sizes of this type of unit.
Fig. 25 shows the distribution from the same unit but with
the reflector satin finished on the inside. It will be noted at
once that there is a considerable decrease in the flux below the
horizontal and an increase in that above, 35.6 per cent, being
within the zero to 6o° zone and 55.2 per cent, below the horizon-
tal. While this is a considerable decrease from that obtained
with the clear reflector, it should be compared with the best of
the opal types, that is to say, the acorn shape where only 24.3 per
cent, of the light, as compared with 35.6 per cent, in this case,
is below 6o°. Where the softer appearance of ground glass is
desired, a unit of this type combines efficiency and such an ap-
pearance.
Considerable variation in distribution .may be obtained in the
case of the reflector-balls shown in Fig. 22 as will be seen in
Figs. 26 and 27 where a 14-inch (35.56 cm.) ball of this type
with a 400-watt lamp was used in both cases but with the center
of the filament in the first case at the center of the ball and in the
6
454 TRANSACTIONS I. £. S. — PART II
second case, 2 inches (5.08 cm.) above. The difference in distri-
bution is quite remarkable and shows that by proper placement
of the lamp, wide variations in distribution can be obtained.
SUMMARY.
From a distribution standpoint, enclosing glassware can be
divided into two distinct classes, one purely transmitting and dif-
fusing, such as ground glass, opal glass, etc., and the other,
prismatic glass, where the principle of specular reflection is em-
ployed. With glassware of the first type, little except good
diffusion and low absorption can be expected, the redistribution
of light being negligible. In the case, however, of prismatic
glass, it is possible to vary the distribution in accordance with the
wishes of the engineer. It is to be noted further that the ab-
sorption of light in both classes is about the same.
The writer regrets that it has been impossible for him to have
conducted depreciation tests on the different units, as this phase
of the subject, namely the loss of light and change in distribution
due to dust, should not be overlooked.
LANSINGH : ENCLOSING GLASSWARE
455
■
^ ^
_, cd ** X
TJ
. u, o ^3 a
<: d i_ 4) is oj
s
OJ
0.0
C
Form
rof filar
e cente
Center
ter of fi
ive cent
as
iJ o
a.
f . - .
"3 - - -
41
u
"
-
-
41 > =2
" o a-2
C .O (J «
0) Cd *"*«.
O N
D. &
5 «
o o o o
0 o >o o
O
in
O
8 8
o o o o o .
>o io o io - o -
3 1
T M
PI PI -H PI Tj-
PI N 10«
00
rO
ON PI
r-» co >- io loso oo
«5
*d- ■* Os On
rO
O 00
PI ON rj- rO lONO O
ro H Tf N
d
M
>i pi
NO rO PI O ONvO NO
ro io !>■ ■<d"
tI-
•+
H Tf
NO 00 vO ON CC - i-i
V J*.
« <J< M
CS
pa
O <
<d < < <j <yj
en
13
• H
(A
u
6
Q H
en
w<5
a
o
CO
c .
4) O
Ocj
CI
Z 55
d*
» .
W o
< <
Q
S
.Q en - - -
41 cd
;
01
13
x'i 1
3 U
■§.«= : : : : :
■S3
a
0)
•343
KJ3
s
3
o M
Ih
o
0)
.3.2
O rf- PI
N
M
I/O PI
Tf" "* PI PI - Tf -
Q.s
p» io o
X
<*
o
00
PI Tt PI PI Tf
«5
oo oo - on
LO
PI
^r _ _ w _ m -
rt TJ-- NO
vO
rO PI On vO - CO "
tO fO PI
cO
IO
M
00 t^ r-» Ov On
s
o
"Alba"
all
"Alba"
all
"cd
XI
<
orn
lilite"
all
id opal
all
id glass
all
" Holophane-
Realite"
Deep " Holo-
phane-Realite "
Holophane
reflector-bowl
Holophane
reflector ball
II
u r
o
U <V XI
cd o
5> --S
5
i4 O
bid
£55
PI rO •'tNO
On
fO •*
vO On H ro lOvO r~»
O
*
—
<n
4J
i-J
3
■oj
0
(0
cri
o
A
O
CJ
u
a,
O
Q
W
it
,_,
<a
W
4)
o
o w
J3
w
od
ca
M
4)
R
00
4)
0)
'J
41
o
^C
00
^4
a
H
o
00
>
s
p
--J
cd
3
cd
5
Q.3
0
H
8
H
N
1
456 TRANSACTIONS I. E. S. — PART II
DISCUSSION.
Mr. H. S. Dunning: In abstracting his paper Mr. Lansingh
brought up the question of the performance of incandescent
lamps in enclosed globes. In the laboratory of the company
with which I am connected, we have been making recently a
series of experiments in which we found it necessary to burn
several of the ioo-watt size tungsten lamps in a small container.
The container was lined with asbestos and there was very little,
if any, ventilation. The experiment which we were making
depended in no way on the maintenance of the lamps, but as a
matter of interest we tested them from time to time and found
very little change. I do not believe that under ordinary condi-
tions the life of the tungsten lamp is materially shortened by
using glassware which practically encloses the entire lamp.
Mr. S. G. Hibben : This paper, contains some very excellent
photometric data on the particular types of enclosing units that
are shown here. But I think I am justified in saying that it is
hardly a fair proposition to give the distribution curves of
totally enclosing units and having selected that particular unit
which gives the most light in a downward direction, to then con-
clude that every time, or under all conditions, this particular
type of enclosing unit is the only one that is "efficient" or justi-
fied in its use.
The conclusion states that with diffusing glassware, little ex-
cept good diffusion may be expected, the redistribution of light
being negligible. This statement will bear considerable quali-
fication. In the first place the positioning of the lamp in the
globe will greatly change the results, and if in these reported
tests the lamps had not been placed so as to give in some cases
the maximum downward reflection, and in others to give dis-
tinctly different distribution, the conclusions could not have
had such an apparently strong foundation. Secondly there is a
change in distribution with a change in shape, even with diffus-
ing glassware. A conical or parabolic shaped reflector made of
white or the so-called opal glass, with a diffusing plate beneath,
would in fact give considerable redirection of light. The above
statement about negligible re-distribution can correctly apply
only to a spherical shaped or a very nearly spherical shaped
ENCLOSING GLASSWARE 457
diffusing globe, of the same quality and finish of glass through-
out.
The user of a diffusing unit may have in mind the attainment
of a re-direction of light, but he also wishes low intrinsic bril-
liancy of the source, a considerable degree of ornamentation, and
possibly a slight color effect. If he wishes downward reflection
primarily, he uses, or should use, an open bottom shade. And if
the enclosing globe meets all these requirements except the first,
it may be used efficiently in the broad sense of the word, and
its use is justified.
I wish to add that, inasmuch as it is the policy of
the society to exclude from the Transactions the special trade
names of various products, when such trade names are excluded,
the proper general name ought to be supplied in every case. I do
not wish to criticise the present paper on this point, but I refer
here specifically to the word "opal' in describing glasses that are
not opal in any sense of the word. In fact, if the two-piece unit
of Fig. i, were actually made of opal glass, as is stated, then
each radius of the photometric curve shown in Fig. 2 would be
reduced at least one-third.
In short, "opal" is not the proper term to apply to the ma-
jority of the diffusing units that have been discussed here. I
would like to suggest in this case to have the proper committee
take up this matter of nomenclature.
The words "opal" and "opalescent" have been used rather
loosely. "Opal" describes a particular class of glass, as does
"crystal," or "alba." There is a sharp distinction between opal
and alba glasses. The opal is a very great absorber of light, and
in some thickness or another will always show a yellowish-red
color of transmitted light. You might use the term "Mazda"
in speaking of all metal filament lamps, whatever the metal or
the burning efficiency, and this would be no further wrong than
to misuse these terms I speak of.
This is a point I wish to bring out very strongly, that the word
"opal" be limited to describing those glasses that are really opal,
and that it be not applied to all white diffusing glasses.
Mr. V. R. Lansingh: (In reply) : Mr. Hibben spoke of the
458 TRANSACTIONS I. E. S. — PART II
possibility of having different shapes in diffusing glassware to
give different photometric curves. It may be possible to do this.
The paper does not attempt to make any comments whatever
as to the use of the glassware which was tested, and consequently
I will say nothing with regard to Mr. Hibben's remark on that
subject, as it is extraneous to the paper itself.
As regards the use of the word "opal," I should be very glad
indeed to have some better word, but at the present time I know
of no more definite classification unless we use the trade names
themselves.
richtmyer: photo-electric cell ix photometry 459
THE PHOTO-ELECTRIC CELL IN PHOTOMETRY.*
BY F. K. RICHTMYER.
Assistant Professor of Physics, Cornell University.
Synopsis: On account of some special peculiarities, the photo-elec-
tric cell presents some interesting possibilities for photometric use. Since
the discovery of the so-called photo-electric phenomenon in 1888, the
process of manufacture of the cells has been so perfected that cells of
sodium, potassium, or rubidium, sensitive to light from the visible spec-
trum, are readily obtainable on the market. Since the intensity of the
current furnished by these cells is strictly proportional to the intensity
of illumination on the sensitive metal surface, the cells may be used for
intensity measurements over a very great range. For the lower inten-
sities an electrometer must be used; for higher intensities a sensitive
galvanometer is permissible. Diagrams of connections, and some sug-
gestions for the several methods of using an electrometer are given. To
use photo-electric cells for photometric purposes however, one must be
perfectly familiar with their peculiarities. For example, the wave-length
sensibility curve lies much farther toward the violet than does the
luminosity curve for the human eye. So that if used for ordinary photo-
metry a specially selected set of absorbing screens must be available. But
in spite of some difficulties attending its use, the high sensibility and the
peculiar action of the cell as a time integrator of light intensity make it
particularly desirable for special photometric purposes.
The photometrist and student of illumination — especially when
working in the research laboratory — frequently meets conditions
which make it desirable to use an apparatus for measurement
which eliminates, in part at least, some of those ever-present diffi-
culties incidental to photometry by the human eye. It may be
desired to avoid various physiological and psychological effects.
Precision greater than that available by eye measurement may be
necessary. The illumination may be of too short duration for
observation by ordinary means. Of the various devices avail-
able for use in such cases none deserves greater attention than
the so-called photo-electric cell. In the hands of one who fully
understands its use and its limitations, it should prove a most
* A paper read at the seventh annual convention of the Illuminating Engineeting
Society. Pittsburgh, Pa.. September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
460 TRANSACTIONS I. £. S. — PART II
valuable addition to any laboratory. It is the purpose of this
paper to point out, in very brief outline, some of its character-
istics, peculiarities, and advantages, and to give several specific
examples for its use.
HISTORY.
Those who have not followed the development of the photo-
electric cell may be interested in knowing that as far back as
1888, Hertz — to whom perhaps more than any other we owe the
wireless telegraph — discovered that under certain conditions a
metal plate, connected to the negative terminal of an electric
generator, the positive terminal of which was grounded, and
illuminated by the ultra-violet light from a spark, would discharge
into the air or to a nearby grounded wire a continuous stream or
current of negative electricity. In the early '90's this interesting
relation between light and electricity was the subject of much
investigation. Among other things, it was shown by Elster and
Geitel,1 that the alkali metals, particularly sodium and potassium,
were sensitive to light from the visible spectrum, especially to
blue and violet. On account of the rapidity with which these
metals oxidize when exposed to the air, it was necessary to study
them in an atmosphere of some inert gas. For this purpose,
Elster and Geitel devised a method, which has since been per-
fected, of pouring the metals, in molten form, into a small glass
bulb containing hydrogen or helium and having the necessary
external electrical connections. These are the photo-electric
cells which may now be obtained on the market in a great variety
of forms for a very reasonable price.
METHODS OF USE.
Fig. i represents diagrammatically the principles involved in
using the photo-electric cell. A battery B has its positive ter-
minal grounded, and its negative terminal connected to the alkali
metal S contained in the cell C. A wire is sealed in through the
glass and connected through a sensitive galvanometer or other
measuring instrument G to earth. If a beam of light be now
allowed to fall on the metal surface in the direction of the arrow
a current of electricity will flow through the galvanometer.
1 See Annalen der Physik, Vol. 43. p. 225 (1891).
richtmyer: photo-electric cell in photometry 461
The strength of this current will depend on several things:
It increases as the electromotive force of the battery B is in-
creased up to a certain point — at least this is true for most cells
obtainable on the market." It depends very greatly on the color
(wave-length) of the incident light. But if these two, i. e.,
electromotive force and color, are maintained constant the
strength of the current is absolutely proportional to the intensity
of illumination on the metal surface over an enormous range of
intensities.3
4hHHH^
ck
#
Fig. 1.— Diagram showing the essential connections for using the
photo-electric cell. B is a battery, its negative terminal con-
nected to the sensitive metal S. The receiving wire or elec-
trode C is connected through a sensitive galvanometer to earth.
The currents furnished are in most cases comparatively small.
Only the higher illuminations (say 20 or 30 foot-candles for
white light) produce currents large enough to be measured by a
sensitive galvanometer, the connections then being essentially as
shown in Fig. 1. These currents are of the order of magnitude
of io-9 amperes. For the more common illuminations (say from
G.0005 foot-candles up) a sensitive electrometer is necessary.
Although the latter instrument is somewhat more troublesome to
handle than a galvanometer, the possibilities of this method of
measurement more than justify its use.4
One may use an electrometer in either of two ways. If the
illuminations are very small the electrometer should be connected
2 When using E. M. F.'s larger than a few volts the' current obtained is due in part to
the ionization in the gas.
3 See paper by the writer, Phystcal Review. Vol. 29, p. 404 (1909).
4 No attempt will be made here to discuss the difficulties incident to the use of an
electrometer. They are by no means insurmountable, and those interested are referred
to the various treatises on the instrument.
462
TRANSACTIONS I. t. S. — PART II
directly in the circuit as shown in Fig. 2. Here the electrometer
Q replaces the galvanometer of the previous connection, and a
key k makes it possible to discharge or insulate the pair of quad-
rants to which it is connected. A condenser M may be used
in parallel with the electrometer to vary its current sensibility
as required. One of the several types of air condensers is most
successfully used.
With this set-up the procedure is essentially as follows : The
illumination to be measured is allowed to fall on the cell and the
key k is kept closed until stationary conditions have been reached.
k is then opened and the rate of drift of the electrometer meas-
Fig 2. — Showing the method of connecting the photo-
electric cell when using an electrometer Q and a
condenser M in place of the galvanometer. A key k
makes it possible to insulate or ground the elec-
trometer.
ured by a stop watch or chronograph. If C is the capacity of
the electrometer system, K its constant (i. e., the number of volts
necessary to deflect it one scale division) and R the measured
rate of drift in divisions per second, then the current I is given by
I = C-K-R
The capacity used with the electrometer should be such that the
rate of drift does not exceed two or three millimeters per second.
However, with a good telescope and scale and proper insulation
and screening for the electrometer rates of drift as low as a few
hundredths of a millimeter per second can be accurately
measured.
This "rate of drift" method has its objections, partly due to
the fact that a comparatively long time is required for each obser-
RICHTMYER : PHOTO-ELECTRIC CEEL IN PHOTOMETRY 463
vation. The writer has used, as a substitute for it in some cases,
the "ballistic" method. This requires in front of the cell a
shutter which can be opened, mechanically or otherwise, for a
definite time interval. In this method, having previously opened
k, the shutter in front of the cell is opened, allowing the illumina-
tion to fall on the cell for the desired time interval, which must
be accurately measured. During this interval the electrometer
has been given a certain charge, in consequence of which a per-
manent deflection is observed after the shutter is closed. If S
is this deflection, T the time of exposure, and C and K have the
same values as before the current is given by
I = C-KS/T
This ballistic method is so useful for such a great variety of
laboratory measurements that a special discussion of some of its
features seems desirable. Experiment seems to show that at
least for moderate illuminations the quantity of electricity dis-
charged by such a cell, subjected to a constant illumination, is
accurately proportional to the time of exposure, even when the
duration of the illumination is only a small fraction of a second.
This makes it possible to make "snapshot" measurements of the
candle-power of a fluctuating light source. The magnitude of
the quantities concerned are indicated by an experiment by the
writer in which a deflection of several centimeters was obtained
by an exposure to a Nernst glower of 0.001 second. The glower
was two or three feet from the cell. It is unnecessary to empha-
size further the possibilities in this direction, for there are
numerous experiments where instantaneous candle-powers are
desirable under conditions where measurement by eye is almost
impossible. Also, one may easily reverse the process and measure
a time interval, as for example the speed of a photographic
shutter.
When used with this ballistic method the photo-electric cell
acts as a time integrator of light intensity, in much the same way
as the electrolytic cell acts as a time integrator for the electric
current. In other words, suppose it is desired to find the average
intensity of a fluctuating light source over a certain time interval,
long or short. One would simply have to expose the cell to the
464 TRANSACTIONS I. £. S. — PART II
light source for the desired interval and read the resulting deflec-
tion. This divided by the time gives the average rate of deflec-
tion. By comparing this rate with that produced by a known
light source the average candle-power of the first source is at
once obtained.
A second method of using the electrometer in connection with
the photo-electric cell is to connect the instrument to the ter-
minals of a high resistance, through which the current flows on
its way to earth. The principle of this method is of course
identical with the method of measuring current by a voltmeter
and resistance. Using this device Nichols and Merritt5 have with
great rapidity measured the densities of a series of photographic
negatives. While this method has the advantage of giving the
current directly by the steady deflection of the electrometer, its
use is limited to cases where fairly high intensities are available.
SENSIBILITY WAVE-LENGTH CURVE.
The manner in which the photo-electric current depends on
the wave-length or wave-length composition of the incident
light is a crucial question when we are considering the appli-
cation of the photo-electric cell to photometry. Numer-
ous investigators seem to agree that as ordinarily used the cell
obtainable on the market has a wave-length sensibility curve
which, while agreeing fairly well in shape with the luminosity
curve of the human eye, is nevertheless quite different in position,
having its maximum much farther toward the violet end of the
spectrum. Fig. 3, I and II, shows the sensibility curve as deter-
mined some time ago6 by the writer for one of the cells then
obtainable. I is the curve as observed without making any cor-
rections, and II is the result of making the correction for the
energy distribution in the spectrum of the source used (curve
III). It is seen that the maximum of curve II occurs at about
0.46 [i, although its exact position is left in some doubt on account
of the difficulty of obtaining a satisfactory curve for the energy
distribution of acetylene, the source used, in the blue and violet.
Later experiments by other investigators seem to show that the
6 Physical Review, Vol. 34, p. 476 (i912)-
6 Physical Review, Vol. 30, p. 3S5 (1910).
richtmyer: photo-electric cell in photometry 465
maximum is even farther toward shorter wave-lengths than here
indicated.
This high sensibility to blue and violet constitutes the chief
objection to the general use of the photo-electric cell for all
photometric measurements. One can readily see that a slight
preponderance of blue in the light as measured by eye, would
make a vast difference in the measurement by the photo-electric
cell, unless the latter measurement were made by use of a very
carefully selected set of absorbing screens, so chosen as to give
20
I5v
10 =
in
! 1
\
2.0
n-ji
I-
m
"t '
■ Y*
|j*S
f\
I \ I
J 1
i \ \l
\
f
1 \ ii
-i.5
/ ■ A
l\ 1 t\
\ 7 *
V
to
1
i
1
i
l\
\ 1
1
m
1 /
, 1
' 1 1
1 7
1 /
fr
I 1
hh-
0.4 y
0.5p .0.6 p
WAVE LENGTH
0.7y
Fig. 3. — Showing the wave-length sensibility curve for a sodium
cell. Curve I is the directly observed data. Curve II is the
corrected curve, correction for the variable dispersion of the
prism used and for the variable energy in the spectrum of
acetylene having been made. Curve III shows the energy dis-
tribution in acetylene (after Nichols and Merritt).
the cell a sensibility curve approximating that of the human eye.
On account of the high absorbing power which such a set of
screens must necessarily have, the sensibility of the device would
be greatly reduced.
For this reason it seems probable that the most useful applica-
tions of the photo-electric cell at the present time must be limited
either to cases involving monochromatic or isochromatic light
sources, or else to investigations where the actual intensity, pho-
466 TRANSACTIONS I. E. S. — PART II
tometrically measured, is of only secondary importance. If, for
example, one were making a series of "stroboscopic" snap-shots
of an alternating current arc over one cycle, he would not expect
the resulting curve to agree either in magnitude or position of its
maximum, with the curve obtained by ordinary means. It is
possible, however, to isolate any wave-length from such an arc
by means of a spectrometer, and by means of the photo-electric
cell follow it during one cycle under conditions of intensity and
color which would render eye measurement impossible. Further-
more, it is to be remembered that in the blue and violet region,
where eye measurement becomes very difficult, the cell becomes
xnost sensitive.
SOME DIFFICULTIES.
In addition to the difficulties previously mentioned, namely,
the great care which must be exercised in studying differently
colored illuminations and also the fact that the electrometer is
somewhat more troublesome to handle than a galvanometer, there
exists one great source of trouble due apparently to the fact that
a photo-electric cell, even in the dark, will allow a small current
to flow if an electromotive force be impressed on its terminals.
This seems to be caused by the walls of the cell possessing a
resistance which is of course not infinite. This leakage current
is fortunately too small to be of serious disturbance when one is
measuring intensities of several foot-candles. But when working
with lower illuminations of a few hundredths of a foot-candle the
leakage current becomes a serious source of error. In fact, it
may become much greater than the photo-electric current itself.
In any event either it must be determined and a corresponding
correction made, or it may be eliminated by charging the cell to
such a potential that the difference of potential between the alkali
surface and the receiving wire is zero.7 in which case there is no
leakage current. Later cells of improved construction have
reduced this difficulty very greatly.
Although the proportionality between photo-electric current
and intensity of illumination seems to hold over a very great
range of intensities, it is probably not safe to expose any cell,
7 See paper by the writer, Physical Review, Vol. 29. p. 71 (1909).
richtmyer: photo-electric cell in photometry 467
for any length of time at least, to the very high illuminations
(several hundred foot-candles). A cell used by the writer, gave
consistent and reproducible results for months when used only
on moderate illuminations, until in one experiment it was exposed
at intervals, totaling perhaps 30 minutes, to a 2,000 candle-power
arc two or three feet away. For weeks afterward this cell gave
very erratic readings when used again on lower illuminations,
the currents being not only not reproducible, but even varying
with a constant illumination, as if to indicate that some sort of
instability had been produced by the very high intensity.
In spite of some serious difficulties, the photo-electric cell is
being used more and more as a photometric device for special
purposes. And the results obtained certainly justify any added
effort necessary for its use.
DISCUSSION.
Mr. J. L. Minick: I should like to inquire as to the per-
manency of the photo-electric cell. If a set of readings were
taken as Dr. Ives has described, and the experiment repeated a
year later, would the photo-electric cell show any deterioration
in that time?
Mr. S. L. E. Rose: I am extremely interested in this paper,
and there are one or two questions that I would like to ask;
I would like to know if the cell is sufficiently developed to be
tried out commercially. The laboratory with which I am con-
nected is ready to help in its application to commercial photo-
metry as soon as the cell has demonstrated its practicability.
Dr. H. E. Ives : I think to Professor Richtmyer is due the
credit of being one of the first to appreciate the really enor-
mous possibilities of the photo-electric cell. He had papers in
the Physical Review I think five or six years ago showing the
laboratory applications of photo-electric cells. I must say that
the photo-electric cell appeals to me as the most interesting sub-
ject in photometry. I have been working on it pretty continuously
for about a year and I feel as a result of that work, the only
criticism I would make of Professor Richtmyer is that he is
not enthusiastic enough about the possibilities, and perhaps he
is not sufficiently impressed with the difficulties of the cell.
468 TRANSACTIONS I. E. S. — PART II
I want to say a word about the sensibility of the photo-electric
cell. Last night Dr. Brashear told us about the eye being able
to look at the sun for half a second and being able to see a
brightness some four quadrillions less. Up to the present time
we have had no instruments possessing anything like that
range. But in recent work by Elster and Geitel the photo-
electric current was measured from illuminations nearly that
from the sun down to that from a pin point before a gas flame
20 or 30 feet away. They claim that a direct proportionality
holds between illumination and current over the whole of this
range. We have something here, therefore, that is going to press
the eye pretty closely for range and sensibility.
Now, as to the accuracy and sensibility of the photo-electric
cell, — I have made measurement after measurement with the
range not exceeding 2/ioths of one per cent., the reading being
just as easy as with a first-rate voltmeter. That certainly ap-
peals I think to people who have had to do much photometric
work.
A great deal of work has been done since Professor Richt-
myer's study upon the theory of the photo-electric cell. It has
been found that the maximum sensibility lies in different parts
of the spectrum for different metals ; with calcium it should be
just where the eye is most sensitive. We have a possibility here
in that cells can perhaps be made with the color sensibility of the
human eye. We might then use the photo-electric cell for
colored light photometry instead of having to call on a large
number of observers or adopt some other roundabout method.
There is a great deal yet to be done with the cell, however.
The question of permanency has been raised and will take time
to settle.
Now in regard to the suggestion of putting the cell in
front of a moving carriage and running it around. If the
speaker would come down to the basement of the laboratory
I am connected with and watch what happens to the electrometer
when somebody sneezes in the room above, he would realize that
there is yet some work to be done before we can carry a cell
around on a moving carriage. Nevertheless, the work I am do-
ing with a spectro-photometer demands extreme sensibility.
PHOTO-ELECTRIC CELE IN PHOTOMETRY 469
Where the light is taken directly, without interposing light ab-
sorbing instruments it has been found possible to use a portable
galvanometer, so perhaps we are not so far from the suggestion
as might appear.
As to the cells being on the market, they are on the market;
but I would not advise anybody to buy them as they are now
being made.
In summarizing this whole question, I want to be very en-
thusiastic. I really believe that for a great deal of laboratory
photometry, especially where lamps of different colors are to be
compared, in a few years we will actually be using the photo-
electric cell in place of the eye, this provided of course that the
simple relationship holds between illumination and current, and
that the outstanding question of permanency and uniformity are
satisfactorily settled.
470 TRANSACTIONS I. E. S. — PART II
FACTORY LIGHTING.*
BY M. H. FX3XNER AND A. O. DICKER.
Synopsis: Ever since the realization of the good and bad effects of
illumination, there has always been a great field in factories for better
lighting conditions. Better light is as necessary as any sanitary require-
ments and with these it should rank among the first. Foreign countries
have taken better illumination a little more seriously than America has.
They have had committees appointed by the government, whose duties are
to study the effects of good and bad light upon the general health and
report upon methods of bettering conditions. Although the importance
of good lighting is generally understood, managers of factories are never
willing to make any decided changes from present operating conditions.
No matter how forcible the arguments, the first cost of the installation
of a lighting system seems to retard any change for better illumination.
We desire to show how easily and cheaply conditions can be bettered. We
believe that it is just a matter of a short time until the factory manager
will understand the great importance of good lighting, and when he does
he will not be satisfied until he has a lighting system that is up to the
minute.
The aim of this paper is to bring out a few of the most
important factors entering into the design or re-design of a light-
ing system for the factory. It is somewhat discouraging to the
illuminating engineer to read article after article dealing with
the methods used to raise the sanitary condition of the factory
and when all has been read he asks himself "What about the
lighting?" Ventilation, cleanliness, devices for safe operation
of machines, rest rooms for employees are all discussed, but
little or no attention is given the lighting. It is the hope of the
authors that this paper will emphasize the fact that factory
lighting is a subject dealing directly with sanitation and that it
should be considered as such. Why is the lighting important,
and whom does it affect? Does it mean a benefit for the central
station only, or is it of equal benefit and importance to employer
and employee? It seems just as reasonable to ask why should a
factory be ventilated or why should it ever be cleaned up. The
* A paper read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
FLEXNER AND DICKER: FACTORY LIGHTING 471
owner or manager would immediately say : "If I do not ventilate
the work rooms the operators will become dull and lose interest
in their work." Regarding his lighting conditions he knows
naught and his answer to a question relative to his lighting con-
dition, would very likely show that he never gave it much thought.
This is just the man who needs some information regarding
lighting. He does not realize that just as many of the headaches
are caused by poor lighting in factories as there are from poor
ventilation. This is not intended to belittle the importance of
good ventilation, but is only mentioned to emphasize the fact that
general improvement of condition does not end when a factory
has been properly ventilated or properly cleaned. It does not end
until the lighting as well as these have been considered. One is
just as important as the other, since injury to the eye from poor
lighting causes suffering equal to or even greater than the sick-
ness caused from poor ventilation. In considering such vital
subjects this country seems to be far in the rear of countries on
the other side. We're behind the times, so to speak, and have
not kept pace with France, England and other European coun-
tries, who are protecting their workmen, along these lines.
In 1912 the French Government appointed a Committee on
Hygienic Aspects of Illumination, composed of prominent physi-
ologists, oculists, engineers, physicists, and inspectors of fac-
tories. The main objects of this committee are:
(a) To study, from the standpoint of general health and its
effects on vision, the various methods of artificial lighting now
used.
(b) To determine the composition and quality, from a hy-
gienic standpoint, of the different combustible illuminants, and
to examine the effect of prejudicial gases and the amount of
heat developed thereby,
(c) To fix a certain amount of artificial illumination to the
normal requirements of vision.
(d) To study the most practical methods of measuring illum-
ination.
(e) To formulate recommendations governing the best means
of applying customary methods of lighting to the chief varieties
of industrial operations.
472 TRANSACTIONS I. E. S. PART II
(f) To present to the Ministry a report on the subject of short
sight and impairment of vision, and on the best methods of
guarding against the cause of myopia.
It is the result of the investigations of such committees that
awaken the mind of the manufacturer to the necessity of pro-
viding good lighting.
The first question that might be asked is : What is good illumi-
nation, or what is practical illumination? Can we spot a unit
or cluster here or there, put a drop light over the working
places in a slip-shod sort of manner and expect to be satisfied
with the results; or, is it a matter of knowing what to expect
from each means of illumination and its corresponding reflector
and to fit in these units to meet the conditions in the factory?
Our common sense dictates that it is the latter. Our experience
teaches us that the problems involved are often difficult of solu-
tion and that we must have definite ideas about correct illumina-
tion before we attempt to accomplish satisfactory results.
One authority defines good lighting as any system which does
not attract attention to the means of illumination, or cause one
to wonder how the illumination was obtained. An analysis of
this yields the following requirements for good lighting:
First, that sources of high intensity must not be in the field
or ordinary vision ; second, that the amount of light be sufficient
for the work to be done; third, that the distribution of light
be uniform or as nearly so as possible, and fourth, that the color
be pleasing to the eye. By adhering to these principles, we will
not go far wrong in laying out lighting installations, whether
for factory or for home, being assured of good illumination.
The value of good illumination should not be under-estimated.
Some are contented to travel along in the old time worn ruts and
to leave well enough alone. Many believe that as long as there
is light, whether good or bad, the question of lighting is settled,
and that the results obtained are as good as any light could
produce. This is the wrong idea, but nevertheless it is enter-
tained by many managers and officials of factories under whose
jurisdiction the question of- lighting comes ; however, they must
realize sooner or later the value of better operating conditions,
produced by good lighting. To do work, light is necessary ; with
ELEXNER AND DICKER: FACTORY LIGHTING 473
a little light, a little work can be done and with more light more
work can be accomplished. This is very evident, and it is easily
seen that no matter how a shop is lighted, if it can be better
lighted, better or more work must result, up to a definite per
cent, increase in efficiency of the workman.*
What if our Mr. Official had to go home to a dimly lighted
dining room? How would he like to read a paper which neces-
sitated straining his eyes, or shave in little or no light, with his
face very near the mirror and his eyes fixed in a staring position.
It would not be very comfortable and he could hardly give him-
self much of a shave; yet under these conditions he expects his
men to work, to turn out good work, and make his factory an
efficient one.
There are such things as good and bad lighting installations,
and to the progressive official the best should not be too good for
his men. However, the initial cost is given first consideration and
is the one stone that lies in the path of all changes, and therefore
we can but sum up the reasons why it is worth every cent that
is asked in making a lighting installation a good and efficient
one.
Statistics have shown that, as the result of better illumination
and a decreased strain on the eye, the physical condition of the
workmen is better, they are better satisfied, imperfections in the
work have been materially decreased and the factory output
increased from 8 to 15 per cent.
Not only is the general physical condition of the workman
improved by better lighting but his liability to accident is greatly
decreased. Recently published statistics show that during those
months of the year in which artificial lighting must be used, there
occurs a greater number of accidents than in the light months.
The saving made by good lighting in this line alone will often
more than repay the extra cost of installing and maintaining the
lighting system. It has been said that a man who is obliged
to keep one eye on the danger points of a machine has only one
eye left to operate it. This is unquestionably true and conse-
quently a machine must be made absolutely safe. The factory
manager usually tries to accomplish this by putting a guard rail
* See Electrical World, page 319, Feb. 10, 1912.
474
TRANSACTIONS I. E. S. — PART II
around the danger points or else enclosing them entirely. This
seems about as reasonable as putting a rail around a hole in a
street without placing a lantern on it. Protected machines still
cause accidents and will continue to do so until the proper light
is provided and the danger points brought well into view. Acci-
dents are becoming more expensive each year and disregarding
all humitarian arguments an owner can no longer neglect to
protect the operators from accidents. Good light is the most
effective protection that can be provided and only carelessness on
the part of the employee will incur accident under these condi-
20 100
It 90
ffO
Z 10
\SOCTL/0ftT
ilUk
^^
idmllll
|! ! ;
J
11
IIIINIIIIII
]llm^t~^^S^~-i
1
I
r
||!!
jjl
y^Da&Hnc-ga 1=4= i = _!*'-'*» Lj ; i - j ■ '
H
|| 111 =H CjLooomess
j~ | | III Gorve:
1 § 1911
| | | III Retort/ I//omi/>otfon.
June July fiuo. «5f/>. Oct. Clov. D£c. </&ri. fee. Mq#. flpfi. fliqt Jem
Chart showing accident and cloudiness curves.
tions. "It costs us a lot of money, but it has paid for itself in
less than a year" said one manufacturer. What more can an
owner want? Certain courts have held that failure to illuminate
danger points constitutes "contributory negligence." Germany,
Austria, Holland and France, realizing the importance of good
lighting conditions, have included lighting in their codes for fac-
tory inspection of health and safety. The accompanying
diagrams shows that the maximum number of accidents occur
during the time in which artificial light is used. It is interesting
to note that the accident curve is almost a duplicate of the cloudi-
ness curve.
FLEXNER AND DICKER: FACTORY LIGHTING 475
In the installation one must take into account the position of
the machines, the work that is done, the location of posts, the win-
dows, and in fact every condition which may in some way
cause deep shadows and bad illumination. The scope of this
paper does not allow us to enter into any detailed account of
layouts. As stated in the introduction, we are only attempting
to emphasize the necessity and advantage of better factory
lighting.
Good factory lighting is not beyond reach ; it is not something
that one can only wish for. It is a material thing and may be
had for the asking.
A great many bad installations can be made good ones by two
inexpensive methods; either re-locating the units and the addi-
tion of proper reflectors, or in some cases by replacing existing
units with some of the modern efficient type now on the market.
It is not hard to show that the new system will, within a given
time, pay for itself, and in a great many cases save money over
the operating and maintenance expenses of the old system.
Assume that the owner of a factory depends solely upon the
profits of the work his employees turn out. An equation express-
ing output must involve the personal equation of the men and
there must be a certain personal efficiency of each man under
every condition in which he works. If a high priced man is
placed under poor working conditions his work will be no better
than the low priced man under good conditions. A manufacturer
will usually buy a labor saving device or a machine with which
his workers can turn out more or better work, and he will supply
his employees with tools of the highest grade steel and have
men to keep these tools in the very best condition ; but he
often absolutely ignores the personal efficiency of the operator
and the conditions under which he must work. He does not
usually see all the methods of making the man as perfect as his5
tools. In other words, more time and thought is given to the
tools than the operator. What good is a perfect tool or machine
if the operator can hardly see what he is doing with it? This
sounds ridiculous, of course, but it is true of many a factory
to-day. For instance, a manufacturer purchased a certain
machine at a cost of $18,000.00 and paid a high priced man of
476 TRANSACTIONS I. E. S. PART II
long experience to operate it. Yet this owner could not see his
way clear to spend $19.00 in order that this high priced operator
would not have to take the product twenty feet away to the
window to caliper it. This shows how little the owner consid-
ered the personal efficiency of his men.
The cost of illumination as compared with an operator's sal-
ary is very small and insignificant; in fact, so small that the
manufacturer can not see it at all. The following data, taken
as average conditions, shows this.
If a 100-watt lamp is assumed for each man and that it burns
3^/3 hours per day for 300 days, the following is derived :
Cost of lamp (Commonwealth Edison Co. renewal) $0.00
Cost of reflector 1 .00
Cost of wiring per outlet 4.00
Total first cost $5-oo
Interest on investment 6% $0.30
Depreciation at 12^ f0 0.63 $0.93
Power at 5 c. 5.00
Cleaning at 3 c. per mo 0.36
Renewal of lamps 0.00
Total $6.29
Wages for 10 hours a day, 300 days, may be assumed to be
$1,000.00. Thus the ratio of the cost of furnishing illumination
to a man under the above conditions would be (overhead expense
not included,), — : , or 0.629 per cent.
1,000
The following mathematical deduction shows what good light-
ing would mean to a factory upon the installation of such a
system. Taking an area of 30,000 square feet with an average
of 0.75 watt per square foot, a connected load of 22,500 watts
would result. Figuring the installation with 250 watt units an
estimate of the first cost is surprisingly low :
90 250-watt outlets at $3.50 $315.00
90 fixtures at $1.25 112.50
90 reflectors at $1.00 90.00
90 lamps (Commonwealth Edison Co. renewal) • . . 0.00
Total I5I7-50
Let it be supposed that this factory turns out a yearly business
FLEXNER AND DICKER: FACTORY LIGHTING 477
of $250,000 and that 33^3 per cent., or $83,333.33 of this busi-
ness is done under artificial light. Assuming a conservatively
5 per cent, increase in output as the benefit due to good lighting,
the business is then increased $4,166.67. If there is a profit of
20 per cent, on this output a credit of $833.33 is derived, which
is considerably more than the installation cost.
As further proof of the low installation and operating costs of
good lighting the following data are submitted from a table com-
piled from actual figures on three trial installations in a large
factory with lamp prices, etc., revised so as to be up-to-date.
100- Watt Tungsten Lamp.
30 reflectors at 92 c $27.60
Wiring at $3. 22 per outlet 96.60
30 lamps at 0.72 21.60
Total $i45-8o
Interest on investment at 6 fo $8. 75
Depreciation at 12^ fo 15-55
Renewals at 30 X 900/1000 hrs. X °-72 *9-44
Energy 3000 X 9°° nrs- X1-1 c- 29.70
Labor (cleaning 30 X 0-63 X 20 c. ) 3.78
Total annual cost $77-22
These figures are derived on the assumption that good factory
lighting will necessitate a 100-watt lamp for 100 square feet of
working area required by an ordinary workman. With these
assumptions the following information has been tabulated :
Total working hours 300 X 10 3000 hours
Total lighting hours 300 X 2>XA IOO° hours
Average cost of labor per hour 35 cents
Labor —
3000 hours at 35 c. $1050.00
Light-
Cost of 1 oo-watt tungsten lamp (Commonwealth Edison
Co. renewal) $0.00
Cost of metal reflector (trade price) 1.00
Average, cost of wiring per outlet 4.00
Initial investment per outlet $5-Oo
47§ TRANSACTIONS I. E. S. — PART II
Interest at 6% 0.30
Depreciation at 12^ % 0.63 0.93
Cleaning 1 2 mo. at 3 c. 0.36
Lamp Renewals (Maintenance) 0.00 0.36
Energy 100 K. W. H. at 5 c. 5.00
Annual operation cost 6. 29
Annual wages for one man $1050.00
Cost of light in per cent, of wages .6
When reduced to cost per hour based on 3,000 working hours
per year, one finds:
Labor per hour $o-35
Light per hour - 0.00629
Cost of light per day 0.02096
Cost of labor per day 3.50
These figures go to show that the cost of good lighting is a
very small portion of the cost for a man's time; in fact, if good
lighting would save five minutes of a man's time per day
a material gain would be experienced.
By following this form, any local conditions causing different
prices than those given can be substituted so that a comparative
figure can be obtained for any particular locality.
The cost of maintenance of tungsten lamps and reflectors is
stated as follows in vol. 1 of the 191 1 Proceedings of the Na-
tional Electric Light Association.
Per cent.
Renewals of lamps 75
Renewals of broken reflectors 3
Changing reflectors for washing 16
Labor for washing reflectors 2
Additional indirect charges 4
100
This data is from experience with an installation of between
7,000 and 8,000 lamps and reflectors.
With the available units, it is impossible to pick out one light-
ing unit and say that it can be used for all conditions. There is
no one cure for all evils. . Individual conditions enter into the
problem and the resulting unit must be best for the conditions
presented. The most important qualifications are the following:
FLEXNER AND DICKER: FACTORY LIGHTING 479
Efficiency; color; quality; arrangement of machines — processes;
adaptability ; special architectural features ; and available hanging
height.
The best unit to use will be the one that best fulfills these
requirements. Each light source, whether gas arc, individual
gas, electric incandescent, arc or vapor lamps, has its definite
field in factory lighting. Usually where one should be used the
others will be less satisfactory. It is hard to convince the owner
that the cheapest is not the best, for he usually wants light only,
and often will not pay for the necessary equipment to produce
illumination. The problem of which one to use depends upon
the class of work to be done under it, as each lamp has certain
characteristics that argue for and against its use.
The last few years have brought great developments in the arc
lamp. The flame arc of long life, furnishes a light source of
high candle-power and low maintenance cost. When the white
light-giving carbons are used the light emitted is of good but
rather variable color. This lamp should never be used in the
normal range of vision. It is best adapted to factories with
high ceilings.
There has been considerable talk about the harmful ultra-
violet rays emitted from arc lamps. These rays are no doubt
given off to a considerable extent, but they are lost in the inner
globe. Therefore this characteristic should not be an argument
against the arc lamp. The greatest objection to this light source
is its unsteadiness, and for fine accurate work a more steady unit
might better be used.
The mercury-vapor lamps are particularly well adapted to
certain kinds of manufacturing. The peculiar color together
with the high visual acuity renders them very useful. A large
clothing manufacturing concern has recently replaced enclosed
arcs with vapor lamps in pressing rooms. It is remarkable
the way scorching can be detected under this lamp while if
a tungsten lamp is used the scorch is not so noticeable. The vapor
lamp has met with decided approval in this kind of work. This
goes to show that the unit used should depend entirely upon the
work to be done.
480 TRANSACTIONS I. E. S. — PART II
In installations where the tungsten lamp is the source of light,
too much emphasis cannot be put on the subject of cleaning.
The manufacturer would not allow his operators to leave their
machines at night without cleaning them; the floors are cleaned
and each morning the factory is found in tip-top shape. Why?
So that the work may begin under the best conditions, all work-
ing toward an increase of output. In other words, everything
but the lighting equipment is systematically taken care of. The
owner knows that the time and money spent in cleaning a machine
is well spent, and yet that which has a greater effect on the effi-
ciency of the operator is left to accumulate dirt from day to
day and in many factories from month to month.
In general, it is best to have the light source as high as
possible above the working plane. If it is out of reach of the
worker, he cannot handle it and thus it will be free of a coating of
oil or other dirt. Truly enough, certain machines require drop
cords in setting up the work or changing the dies, but few ma-
chines actually need drop cords during their operation. One
big railroad shop in Chicago has adopted Cooper-Hewitt lamps
for general lighting and drop cords are checked as any other
tool. In this way they are taken care of and are not used except
when necessary. It has been our experience that the worker
will use a drop cord as long as he has one in front of him.
The first move for efficient lighting is general illumination,
where possible, doing away with the drop cord or as above stated,
making the drop cord a working tool.
There are many combinations of efficient lighting systems; in
fact, it is a subject of its own, so that we will not attempt a dis-
cussion.
Realizing the general disregard of good lighting as a necessary
and important part of the factory equipment, and not overlooking
the attractive lighting load of this class of service the Common-
wealth Edison Co. of Chicago decided to make a proposition cov-
ering lighting installations for factories.
As has been stated, the first cost of the installation is too often
the only obstacle, and therefore this company decided that the
first way to make such a proposition attractive was to do away
with the first cost. To insure most efficient operation the com-
JXExner and dicker: factory ughting 481
pany includes in this proposition the cleaning and renewing of
all fixtures and lamps.
The Commonwealth Edison Company's proposition is as fol-
lows:
The customer is asked to sign a contract for a period of
twenty-four consecutive months. After the expiration of this
period, the wiring and fixtures become the property of the con-
sumer.
Charges for this service are made up on the following
basis : rental charge, maintenance charge and electricity charge.
Rental Charge. — The rental charge is twenty-five cents per
fixture per month, allowing the consumer to use either 100, 150
or 250-watt units in each fixture. At the end of the two year
period this equipment becomes the property of the consumer and
this charge is discontinued.
Maintenance Charge. — The consumer pays the company twen-
ty-five cents per fixture per month, except during the months
of June, July and August. At the end of the two year period,
the consumer may elect to discontinue paying this charge and
take care of this equipment himself.
Electricity Charge. — For this service the consumer pays our
regular schedule A rate which is ten cents net per kilowatt hour
for the first thirty hours use of the maximum demand per month
and five cents net per kilowatt hour for all energy used in excess
of this amount.
The fixture supplied under this contract is one that was es-
pecially designed for this class of service. It consists of a
shallow reflector with a collar containing a lock socket ; the
conduit serves as a stem. The reflector is so designed that the
filament does not extend below the bottom of the reflector.
Photometric curves show extensive characteristics. The idea
throughout was to make a reflector that was efficient, plain, and
easily cleaned.
The Commonwealth Edison Company confidently expects to
install 10.000 of these fixtures within the next year, and a re-
port of the first few months gives reason for the confidence
expressed.
In summing up we believe that the campaign for good fac-
482 TRANSACTIONS I. E. S. — PART II
tory lighting has just begun and that the best argument in favor
of better illumination is a statement showing the benefits derived
from an efficient lighting system and the experiences of others.
Even a hasty reconsideration of the arguments presented in
this paper demonstrates the tremendous scope and possibilities
along this line. There is no longer any excuse for poor light-
ing; the necessity, the practibility, and the economy of good
illumination have been demonstrated beyond question and if
the strides in this direction which have been made in the recent
past may be taken as an index of those which will be made in
the future, there is no doubt that very soon the time worn phrase
"a badly lighted shop" will have disappeared from the vocabulary
of those connected with the lighting industry.
We believe that if a fair and broad-minded manufacturer will
but figure out in a common sense way the merits and necessity of
good illumination, he will be converted to its use in a short time.
If his own figures do not satisfy him, let him consult those who
have been far-sighted enough to go ahead with his better sense
dictation and be shown, if necessary, the truths of the above
assertions. He will realize sooner or later the needs of his men
— better atmosphere, lighter and cleaner shops, and proper illu-
mination.
discussion.
Mr. H. A. Reid : On the seventh page of the paper there
seems to be a discrepancy with regard to the assumed cost of
wiring per outlet, the reason for which is not apparent. In the
first table the cost for 100-watt lamps is given as $4.00, and in
the second table $3.50 is used as the cost for 250-watt outlets. It
would seem that the cost of wiring should be no higher for the
100-watt than for the 250-watt; rather should the reverse be the
case. In the actual installation cited on the eighth page the cost
per 100-watt outlet was $3.22.
The assumed depreciation rates appear also to vary. In the
first example on the seventh page, 12.5 per cent, is taken to ap-
ply on lamps, reflectors and wiring; in the second example the
rate on wiring is reduced to 5 per cent. If the wiring is prop-
erly installed, the latter value will probably be about right.
FACTORY LIGHTING 483
Mr. R. B. Ely : I think the central stations and gas companies
particularly the smaller companies, should encourage the intro-
duction of illuminating engineering departments. The up-to-
date companies can effect a saving and improve the lighting con-
ditions in factories, stores, etc., and of course where there is
gas or other means of illumination employed, the advice of the
engineering department when followed will frequently offset the
expenses of the improvements. In numerous instances it takes
a week to lay out an installation. In one case it cost $1,200 to
equip one floor for trial, but the effect and saving was so good
that they went ahead and wired the whole factory. That is not
an exceptional instance.
Mr. G. H. Stickney: I want to emphasize the importance
of industrial lighting to the central station. I have felt in the
past that a great many central stations have considered indus-
trial lighting as an undesirable load and have not awakened to
the possibilities of it. I think that now that the Commonwealth-
Edison Company, one of the most enterprising companies of the
country, has pointed out what they are doing, we may hope that
in another year other central stations may also contribute useful
data.
Mr. G. W. Roosa : In the second paragraph on the tenth page
is this statement: "When the white light giving carbons are
used the light emitted is of good but rather variable color." In
factory lighting where flame carbon arc lamps are to be used the
bays are usually wide and high. Such installations often include
a number of these lamps ; for instance, four or more, or possibly
a multiple of four. If there is a variation of candle-power from
any one lamp it can not be altogether objectionable because the
average variation from several lamps will be slight, due to the
fact that variations in individual lamps seldom occur simul-
taneously. Flame carbon arc lamps are usually equipped with
diffusing glassware of some kind when. placed within the field of
vision, or at a position approximating it. Equipped with proper
diffusing glassware this type of lamp has an intrinsic bril-
liancy of about 15 candle-power per square inch. On the other
hand, most incandescent lamps are used with clear bulbs instead
484 TRANSACTIONS I. E. S. — PART II
of frosted ones, and I believe the intrinsic brilliancy is about
1,000 candle-power per square inch.
Mr. E. W. Lloyd : It occurs to me that the benefit to be de-
rived by the manufacturers, central stations and illuminating en-
gineers from the development of good lighting in factories would
put more money in their pockets than from any other line. We
have heard a great deal about the developments of motor drive
in central station service in factories all over the United States.
The recent United States census shows some very remarkable
figures relative to that growth — several hundred per cent, in 10
years. I believe the growth will be greater in the next 10.
Strange to relate, however, the development in factory lighting
has been very, very, slow ; at least it is slow in many of the cities
that I am acquainted with. There are many motor equipment
in factories, but an examination of the installations of lighting
in those factories is a shock to a man who spends any time think-
ing about the subject of illumination. It was because of that
condition in Chicago that we decided to help increase this busi-
ness by offering to finance the installation of these lighting sys-
tems.
Manufacturers are becoming interested in the proper design
of reflectors and fixtures, and I believe there is a great deal
being done in this direction. All manufacturers of fixtures are
not up to date. Some of our trouble comes from the different
schools of illumination, the different methods of accomplishing
results; but I don't think it is necessary to dwell upon that.
Every effort toward better illumination is going to help the whole
matter along.
I merely want to say in closing that the central stations can
do more business at the moment, by devoting energy to the illum-
ination of factories, in my judgment, than in any other way.
Mr. H. C. Chapin : It is important to emphasize that tests
be made of working conditions ; for instance, a yellow flame
carbon, already of higher efficiency than a white flame, will show
a still greater comparative efficiency in a , smoky atmosphere,
because the yellow light penetrates the atmosphere better than
the white.
Mr. S. L. E. Rose : It may be of interest to know that there
FACTORY LIGHTING 485
have recently been developed reflectors for the flame arc, which
give intensive and extensive distributions and also an angle re-
flector. A 6.5-ampere direct-current, multiple, flame lamp with
yellow carbons and an intensive reflector gives about 10,000
lumens in the zero to 60 degree zone. This is equivalent to about
14 lumens per watt, and applies where the lights necessarily
hang high. The extensive reflector on the same lamp gives
approximately the same number of lumens in the zero to 60
degree zone.
Mr. M. H. FlExnEr: I wish to say that our information as
to the increase in output, resulting from better lighting condi-
tions, was obtained from a pamphlet on factory lighting by
Mr. G. C. Keech and from the plant of Wilcox and Parker of
Chicago. The latter firm reported about a 10 per cent, increase
in output.
Our remarks on the flame arc lamp were based upon our
own observation. Flaming arcs should not be hung low, and by
that I mean hanging heights of 10 and 12 feet. In such in-
stallations bright spots are plainly visible, proving that the lamps
are not lighting as large an area as they would with higher
hanging. The City of Chicago has recently passed an ordinance
which requires that all arc lamps in the down town district be
hung 23 feet, and in the outlying districts 21 feet. There are
some streets lighted with flaming arcs and which are hung
about 10 feet high. In riding in the street cars on those streets
one finds the glare from these lamps very disagreeable, and
riding through the alternate dark and light spaces, soon fatigues
the eye. This, of course, is a little foreign to our subject, yet
the results would be somewhat similar in both installations re-
ferred to.
The discrepancy in costs of wiring per outlet, given on the
seventh and eighth pages, referred to by Mr. Reed is due to
the fact that one is an actual and the other an average cost. The
discrepancies in depreciation, etc., are explained by the fact that
conduit was used in one case and moulding in another.
Mr. H. H. Magdsick: This paper brings to our attention
the very low cost of lighting compared with the total cost of
production in industrial plants. Recently we obtained com-
8
486 TRANSACTIONS I. E. S. — PART II
plete data for a cold-roll mill lighted uniformly to an intensity
of two foot-candles, which made it possible to read micrometers
with ease in every part without any auxiliary drop lamps. The
total cost of operating the lighting units, including the fixed
charges on the equipment, renewal and maintenance of lamps
and the energy, is only one one-hundred-fiftieth of the cost of
labor during the hours of darkness. In other words, supply-
ing all the light that is desired in this department costs no
more than furnishing one helper for each 150 men. The
difference in cost between poor illumination and adequate light-
ing is, of course, still less. Any system that would enable
this mill to be operated at all would, we will say, require one-
half the expense; the difference between poor lighting facili-
ties and the very best illumination is, then, secured at a cost
no greater than that of supplying one helper for 300 men.
And the cost of labor is only one item in the cost of pro-
duction, often no more than 20 per cent, of the total. If to
it are added the fixed charges on the plant and machinery, the
cost of power and expense of supervision, administration, etc.,
one obtains an idea of how comparatively insignificant is the
cost of proper illumination.
The iron and steel industry offers an interesting example of
recent developments in plant illumination. President Parkhurst
of the Association of Iron and Steel Electrical Engineers has
given permission to quote from a paper that will be presented
before the convention of that association now in session. The
data contained therein were compiled from reports submitted
by nearly one-half of the companies represented in the A. I. &
S. E. E. It will be recalled that the members have for several
years interested themselves actively in illumination matters and
were among the first to co-operate with this society.
From the reports, it appears that during the past two years
the wattage installed in lighting equipment throughout the plants
has decreased slightly, about 5 per cent, but the average inten-
sity of illumination has been raised to about 135 per cent, of
the former value. This increase was secured by replacing obso-
lete equipment with more modern units. Thus, while two years
ago about 58 per cent, of the connected lighting load was in
factory lighting 487
carbon arc lamps, for the most part of the 220-volt type, and
less than 8 per cent, was in high-efficiency incandescent lamps,
the carbon arc lamps have now dropped to about 34 per cent,
and the high-efficiency incandescent units have been increased
to an equal wattage. The carbon incandescent lamps have not
been decreasing as rapidly as the arcs due, it appears, to the
fact that up to the present time the intensities of general
illumination provided have proved insufficient to warrant dis-
continuing the use of drop lamps for local lighting. Where
complete modern systems have been installed, however, the
reports show that auxiliary units have in practically all cases
disappeared. These changes represent the important develop-
ments in the lighting equipment of the steel mills. The flame
arcs, mercury-vapor lamps, etc., remain a small percentage of
the total.
According to the data submitted, the 76 plants represented
in the A. I. & S. E. E. still have about 24,000 kilowatts
connected in carbon arc and carbon incandescent lamps. If
these were replaced with modern equipment, the illumination
instead of being 135 per cent, of the average value of two years
ago, would become 235 per cent. It is costing the plants about
$1,500,000 annually to operate this obsolete equipment. At a
conservative estimate, fully two-thirds of this amount, or
$1,000,000, is wasted annually in view of the higher efficiency
of the modern equipment with which it might be replaced.
The fact that when changes are made this wasted amount will
probably be expended in the operation of modern lighting sys-
tems in order to secure higher intensities of illumination, serves
merely to multiply the loss, for the gain in the output of the
mills and increased safety of employees is in practically every
instance far greater than the cost of illumination.
488 TRANSACTIONS I. £. S. — PART II
HOSPITAL LIGHTING.*
BY WIWJAM S. KILMSR.
Synopsis: Short-comings in hospital lighting are too well known to
need mention; ordinary illumination devices for hospitals fail to give the
proper quality and distribution of light. This paper treats solely of the
lighting of the two most important quarters of the hospital, viz., operating-
room and wards, and describes practical fixtures for the solution of the
various problems. Although written from an engineering standpoint the
paper is based on intimate knowledge of hospital conditions derived from
the author's two years' concentrated work in this field.
The modern hospitals of to-day are undoubtedly models of ele-
gance and hygienic forethought and this high degree of perfec-
tion has been reached by experience in overcoming specific diffi-
culties and conditions as they are encountered.
For some unexplainable reason the question of proper artificial
illumination has received little or no consideration, thus greatly
lowering the efficiency of the splendid service characteristics of
these institutions.
Light like any other form of energy may become an agent of
destruction or a minister of health, precisely in accordance with
the wisdom shown in its application, and it is the duty of the
professional advisor, whether he be architect or engineer, to
understand these particular conditions before he can pretend
to satisfactorily specify a lighting system.
The operating rooms and wards which are the most important
parts of a hospital naturally present the most difficulties, and it is
the object of this paper to treat particularly the lighting of these
two areas.
OPERATING ROOMS.
The operating table should have, on account of the large amount
of surgical work performed at night, a system capable of con-
centrating on the field of operation an illumination intensity of
not less than 25 foot-candles, preferably higher, and approximate
* A paper read at the seventh annual convention of the Illuminating Engineering
Societ3', Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
KILMER: HOSPITAL LIGHTING
489
the tonal value of daylight. The equipment should be designed
along plain smooth lines, thus avoiding the septic risks of dust
collections and rendering it easily and thoroughly cleaned. When
it is necessary to suspend a fixture in close proximity to the
table and when over 250 watts are consumed by the equipment,
Fig. 1. — Ventilated operating fixture, using six ioo-watt tungsten lamps.
it is most important that the heat generated by the lamp be
directed away from the patient and the head of the surgeon.
Fig. 1 illustrates a unit of the above description, which is de-
signed for six 1 00- watt tungsten lamps and when suspended
6 ft. 6 in. (1.98 m.) from the floor to lower edge of the frame,
49Q
TRANSACTIONS I. E. S. — PART II
an average illumination of 40 foot-candles is distributed over the
table. By the addition of a ball and socket joint this equipment
may be adjusted to suit any form of operation.
To reduce the heat generated by the lamps to a minimum, the
fixture has an ingenious arrangement of double glass slides, by
Fig. 2. — Ventilated operating fixture, using eight 35-watt double base tubular lamps,
with emergency gas attachment.
which a forced draft is created by the heating, and consequently
raising the air between the glasses, the heated air passing out of
the fixture through the vents at the top. This syphon arrange-
ment is often assisted by the use of suction pumps.
Under actual working conditions, temperature tests show the
KILMER: HOSPITAL LIGHTING
491
following results, with mercury 9 in. (22.86 cm.) below the lower
plate glass :
Degrees
Fahrenheit
Temperature of room 7&
After one hour burning with slide 83
After one hour burning with no slide 94
Fig. 4 shows an equipment of this character installed in the
Post Graduate Hospital, New York City.
Fig. 3. — Detail of dustless indirect ward fixture.
Fig. 2 is a unit built on the same general principles, but designed
for a double base tubular lamp with a straight filament. This
fixture with eight 35-watt lamps gives an illumination equal to
the previously described unit, but over a smaller area. The
elimination of the excessive heat is as follows :
Degrees
Fahrenheit
Temperature of room • • 78
Fixture after one hour burning with slide 81
Fixture after one hour burning witii no slide 87
492 TRANSACTIONS I. E. S. — PART II
In both of these fixtures the mirrors forming the reflecting
surface are so arranged that the field in front of the surgeon's
hand is always free from shadows, no matter in what position
it may be.
The new concentrated filament lamp, with a distribution of
light given in the following table, has made possible another
form of lighting the operating table, vis., adjustable reflectors
capable of powerful concentration. (Electrical Testing Labora-
tories report on ioo-watt concentrated tungsten-filament lamp
referred to) :
Angles
Candle-
degrees
power
155
66.O
145
72.0
135
75-o
125
75-5
115
76.0
105
76.5
95
74.0
90 Horizontal
74.0
85
76.0
75
78.0
65
80.0
55
81.0
45
82.0
35
78.0
25
72.0
15
64.0
5
56.0
0 Nadir
55-o
Fig. 6 is an all metal reflector so designed that the light is con-
fined approximately to an area 50 either side of the vertical and
should be suspended stationarily directly over the table, about
8 ft. (2.44 m.) from the floor. When it is so installed with a
1 00- watt concentrated filament lamp a satisfactory distribution is
given for all operations requiring a powerful downward distri-
bution. To protect the eye of attendants and surgeons, when
raised from the field of operation, a series of metal bands may
be inserted as shown in the drawing.
The following photometric values apply to this unit:
^
Fig. 3a. — Exteiior of dustless indirect ward fixture.
Fig. 4.— Fixture shown in Fig. 1 installed over operating table in
Post Graduate Hospital, New York City.
Fig. 5. — Adjustable parabolic reflector for powerful concentration.
Fig. 6. — Concentrating reflector for use directly over operating table.
KILMER: HOSPITAL LIGHTING 493
Apparent Candle-power
Angles ' 1
Degrees Without bands With bauds
o Axis 2,160 2,160
5 2,250 j, 800
10 430 380
Photometric distance 30 ft. (9.144 m.) ; measurements made in
a single plane through the reflector axis.
Abdominal and pelvic operations require a penetrating beam
of light. Fig. 5 shows a parabolic all metal reflector 18^5 in-
(47.59 cm.) in diameter, adjustable to any angle. This equip-
ment should be attached to a rigid support about 12 ft. (3.66 m.)
from the floor and 12 ft. to 15 ft. (3.66 to 4.57 m.) to the rear
of the table with the adjusting rod of sufficient length to be easily
reached; when so installed a powerful beam of light is directed
over the shoulder of the surgeon directly into the field of
operation.
The photometric values of this unit are as follows :
Angles Apparent Candle-power
Degrees , — : • ,
0.0 9,600 7.940
2-5 8,330 —
5.0 2,000 —
Photometric distance 30 ft. (9.144 m.) ; 100-watt concentrated
filament tungsten lamp, measurements made in a single plane
through the reflector axis.
It is quite apparent that with this distribution there need be
no fear of insufficient and poorly distributed light. These equip-
ments may also be mounted on adjustable standards, thus having
an equipment for all emergencies which are so often encountered.
Some hospitals object to any suspended lighting equipment in
the operating room; this usually applies to rooms having a sky-
light, and provided sufficient head room is available a satisfac-
tory system may be designed by placing a series of reflectors
around the edge of the skylight. These reflectors may be so
designed that the maximum reflection falls at a point directly
over the table. This scheme is likely to.be more satisfactory than
an attempt to illuminate the entire skylight so as to provide a
sufficient amount of light for an operation in any part of the
room. The latter scheme would necessitate a very large con-
sumption of energy.
494 TRANSACTIONS I. £. S. — PART II
WARD LIGHTING.
In the wards the evils of glare have to be most carefully
avoided, as influence of glare upon the retina of the debilitated
or depressed is usually followed by serious results.
The average ward of twenty beds should be provided with two
kinds of illumination, general and localized. The general light-
ing should not average more than one foot-candle and should be
obtained from either a totally or partially indirect source, so
arranged that violent contrasts are entirely eliminated. It is
advisable that the ceiling be treated with a flat white or French
zinc surface and the walls in a slightly darker tone, such as
buff or French gray.
The fixtures must be of special design and the same hygienic
conditions apply as in the operating room, vis., elimination of
dirt and heat. Also that portion of the light which is diverted
to the ceiling must be evenly distributed without spots and
streaks. The exterior finish should be a restful color such as
green enamel or matt nickel. This will not give a violent con-
trast, and will overcome the sameness of the white enamel finish
which many fixture manufatcurers seem to think is necessary
for hospitals.
Fig. 3 shows a form of indirect fixtures which has met with
great success in ward ligthing. It can be made for any num-
ber of lamps up to six. The interior is lined with a series of
mirrors set at any angle to insure an even distribution of light
over the ceiling. The angle at which these are set depends on
the distance between outlets and the size of the ward. Over the
top of the bowl is set a thin blown glass plate which may be
easily raised for lamp renewal by means of the sliding metal
holder, and which renders the interior of the fixture dust and
dirt proof. Ventilation is accomplished by means of vent holes
at the bottom where the air enters and passes through and es-
capes between the housing of the fixture and the glass plate. A
view of these fixtures installed in one of the wards of the Post
Graduate Hospital, New York, is shown in Fig. 7.
If it is so desired one lamp in a fixture may be arranged
to operate on a separate switch, but this is not advisable on
account of its effect on the patients. Night lights, wherever
/
VI H
Kig. 7. — Night view of typical ward in Post Graduate Hospital. New York,
using dustless fixture shown in Figs. 3 and 3a.
ft
^_
0
~y^y~
Figs. 8 and 8a.— Adjustable ward bracket, lower figure showing
correct method of installing.
Fig. 9. — Conical microscopic reflector.
KILMER: HOSPITAL LIGHTING 495
necessary should be in the form of brackets and as inconspicu-
ous and as far out of the range of vision as possible.
When a direct indirect system is to be used the character of
the glassware used for the bowl or hemisphere must receive a
thorough study ; it should be of sufficient density to maintain the
presently given intensities per square inch of exposed sources and
have diffusion qualities which render it impossible to have any
bright spots or streaks from the lamps exposed to the eyes of
the patients. It is also advisable to eliminate the necessary gas
emergency equipment from these fixtures, but it may be included
in the localized equipment.
For the localized lighting of the ward the two forms most
advisable are proper brackets or "bedside" lights ; everything
depends on the word "proper." No source of light or illuminated
surface exceeding 0.05 candle-power per square inch (7.2 candle-
power per square foot) should be exposed to a patient's eye;
therefore, the usual direct lighting equipment is quite out of the
question. An excellent form of bracket fixture is shown in Fig.
8. It is adjustable in every direction and it is equipped with a
reversible glass reflector with opaque sides. The reflecting por-
tion is of a concentrating type, designed especially for bedside
examination, thus eliminating any necessity of the use of a special
portable lamp by attending physicians ; and by reversing directly
upward this concentrated light is directed to the ceiling, thus
giving an excellent form of indirect illumination.
The opaque sides render it impossible for a ray of light to
reach the eye of a patient in any part of the ward and enable
any bed to be immediately illuminated without any disturbance to
other patients. An entire ward equipped with one of these
brackets to each bed would not require any other form of illum-
ination and the emergency gas equipment may be easily incor-
porated in this fixture. Many hospital equipment specialists and
authorities predict this to be the future form of ward lighting.
"Bedside" lights are the least desirable for a localized equip-
ment, but where absolutely necessary, they should be carefully
designed. The reflector portion should preferably be hemis-
pherical in form, opaque and mounted on a plain standard
heavily weighted at the base. The reflector should be adjustable
496 TRANSACTIONS I. E- S. — PART II
by means of a knuckle joint. The reflecting surface should give
a diffused reflection as from opal glass or flat white enamel.
Of course an additional light is required for the physician for
bedside examinations, thus making this equipment more costly
in both initial expense and maintenance, as both equipments have
to obtain the energy from base board receptacles which naturally
necessitate a considerable portion of exposed cord, making the
liability of breakage considerable and costly.
In the examination of bacteria under a powerful microscope,
it is always difficult to arrive at definite results when observers
at various points are working under light sources of varying
spectral value. Varying light sources naturally produce a dif-
ferent tonal value to a germ so that it is often difficult, if not
impossible, to be sure of identification.
Fig. 9 shows an instrument whereby there may be obtained
at any time a uniform light source giving approximately the
effect of a slightly veiled north light, which is so much desired
in precision work, and far superior to the average illumination
obtainable from city windows. The light from a 60-watt all
frosted tungsten lamp operating at exact line voltage is directed
by means of a double convex reflector with the lower surface
of silvered glass and the upper surface of opal glass.
A carefully calibrated liquid lens is inserted at the aperture.
After the light has passed through this lens the resultant illum-
ination on the microscopical field is of a daylight spectral value.
For dark field work this lens is changed to one giving an illum-
ination of a powerful violet character.
Other lighting apparatus of the hospital has been intention-
ally not discussed in this paper because in the author's opinion
it does not call for the study given the conditions herein men-
tioned.
The lighting of halls and corridors is in a class with the
lighting of hotels and office building corridors. The nurses and
doctors quarters should receive the same careful treatment
required for the lighting of the home. The dispensary should
be given the same treatment as the modern drug store; particu-
lar attention should be given to the lighting of the prescription
counter.
HOSPITAL LIGHTING 497
DISCUSSION.
Mr. H. Calvert: This paper is a good article for a person
to read who is about to design the lighting of a hospital. It is
very practical. The lamp fixtures shown for the operating room
certainly would give good results ; but one reason why the author-
ities of some hospitals object to having the lamps placed directly
over the operating table is the liability of dust or dirt falling
from the fixture and infecting the patient. To obviate this, some
hospitals have adopted the plan of installing four units approxi-
mately above the corners of the operating table.
Referring to ward lighting, indirect fixtures are quite suitable
provided the amount of illumination which falls on the ceiling
is not excessive, as the patients, a large part of the time, are
lying on their backs and to gaze upwards at a very bright ceiling
is of course uncomfortable. The adjustable ward bracket which
is shown in the paper is a good one for the purpose. In one of
the Philadelphia hospitals this bracket arrangement has had a
unique feature added by equipping the bracket with a pull socket
so that the patient can light the lamp, and at the same instant a
little pilot lamp which is located near the desk of the head nurse
is also lighted indicating, if she has not already observed it,
that the patient desires attention.
Mr. W. F. Little : Referring to the lighting of operating
tables, I note that Mr. Wheeler states that an intensity of 10-
foot-candles has been found sufficient. Comparing this with il-
luminations given by the Zeiss* system, we find intensities in
the neighborhood of 300-foot-candles, and I further understand
that this system has been used with very good success in various
hospitals. I should be interested in knowing whether Mr.
Wheeler believes 10-foot-candles sufficient for operating table
lighting in general.
Admittedly, the light of hospital wards should be done under
the principle of the greatest good for the greatest number.
Therefore, would it not be feasible to -turn an indirect system
of illumination upside down directing the light first upon the
white floors and from there diffused throughout the room?
* Illumination measurements of Zeiss system given in I. E. S. Transactions, June,
1912.
498 TRANSACTIONS I. E. S. — PART II
Mr. H. B. WhEeX£r: Operating Rooms. The lighting of
hospitals is a very important subject and one we know very little
about. This is especially true in operating room illumination.
It has been my experience that extreme concentration of light
on the operating table, not only gives a multiplicity of shadows,
but an intense heat, both the opposite to what is desired, namely,
an evenly diffused light. Indirect Illumination from the ceiling
and walls gives an abundance of diffusion eliminating completely
objectionable sharp shadows.
Several large operating rooms of the Toronto General Hos-
pital and St. Mary's Hospital at Rochester, Minn., have been
lighted for some time with indirect illumination which is giving
very satisfactory results for operating purposes. The indirect
fixtures are very plain arm chandeliers with adapters for sup-
porting one piece silvered glass reflectors. The average inten-
sity is approximately 10-foot-candles.
Another system of diffuse illumination has been in use for
the past six years in the Southern Pacific Hospital at Los An-
geles, Cal. It consists of a battery of one piece silvered glass
reflectors suspended over a chipped glass skylight. In installing
a system of this character, care should be taken to select glass
of good diffusing qualities.
Corridors. Corridor lighting is just as important as the light-
ing of any other room in a hospital, because patients are con-
tinually being taken from the wards and private rooms to the
operating table at critical stages. For this reason the lighting
should be concealed as in other parts of the institution.
Wards. It is very desirable in wards to have a flexible sys-
tem of indirect illumination.
A low intensity {%. foot-candle) for a night light, a medium
intensity (y2 to 2 foot-candles) for reading, etc., and a high in-
tensity for close examination of patients are required. Local-
ized direct lighting portables attached to baseboard outlets be-
tween beds are used largely for the high intensity.
Generally in large wards varying degrees of illumination
are controlled by electrolier wall switches and the smaller wards
by switches on the fixtures.
When luminous bowls are desired, the small lamp for illumi-
nating the glass bowl may be used for a night light.
philbrick: store; lighting 499
STORE LIGHTING.*
BY J. E. PHILBRICK.
Synopsis: The gaslighting installations of eight small stores are out-
lined in this paper. Plans showing the locations of test stations, and
illumination readings of each store are also included. The author em-
phasizes two points: (i) gaslighting solicitors should have available for
prospective customers accurate data on local lighting installations; (2) the
need of proper maintenance to insure maximum efficiency and satisfaction.
The author hopes that the publication of the tests and data
of this paper and a discussion of them may go far towards elim-
inating the old "hit or miss" methods used by commercial depart-
ments in securing business and also in making installations.
The following table gives average resultsf of tests of the light-
ing installations in several stores, plans of which are shown in
the accompanying illustrations.
Table 1.
lumens per
cu. ft. of gas per
hour light ceiling
Nominal eon- With With
sumption of light dark
fighting unit unit cu. ft. walls walls
Reflex lamps with frosted tip cylin-
ders, prismatic or light (imported)
opal concentrating reflectors t>1A 125 114
Reflex lamps with frosted tip cylin-
ders, prismatic or light (imported)
opal distributing reflectors 3^ no 100
Reflex lamps with frosted tip cylin-
ders, French roughed ball globe-.. t>xA 95 70
Reflex cluster lamp, four-mantle, with
alabaster globe 13 S5 64
Inverted five-mantle arc with alabaster
globe 16.6 87 65
Upright four-mantle arc with opal re-
flector and alabaster globe 20 75 55
Upright four-mantle arc with alabaster
globe only 20 66 48
* A paper read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
t Figures published by a manufacturer of lighting appliances for use in the design of
gas lighting installations.
500 TRANSACTIONS I. E. S. — PART II
The installations tested were selected at random and under
ordinary working conditions, no special preparation of lamps or
mantles being made in any case. If renewals had been made they
were made at the ordinary time for cleaning, and as the test
shows there is but one case in which the units had received atten-
tion within one day of the time of making tests.
For some time the opinion has prevailed that figures relative
to the performance of gas lighting units required radical dis-
counting in order to express the actual results secured upon a
consumer's premises, under the conditions of care and attention
usually encountered in actual service.
Obviously, it is most important for the man responsible for
the success of any business to obtain accurate information con-
cerning the excellence of the service rendered by his product
with particular reference to those features directly under the
observation of the consumer, and upon which the latter bases
his judgment of the service and product. Many gas companies
have been deterred from gathering this data by the effort re-
quired to obtain it. I believe, however, that the value of this
information is amply demonstrated in the tests reported in this
paper which were made in York, Pa. Tests of this character
not only enable the manager to determine the competitive posi-
tion of his product and to keep a check on the capacity of his
manufacturing distribution and maintenance departments as re-
gards the performance of their various functions in supplying
lighting service, but form a basis for calculation by salesmen in
designing installations and advising customers.
When salesmen use data furnished by the manufacturer of
lighting appliances, they are naturally inclined to make rather
liberal discounts to allow for discrepancies between service and
laboratory conditions, and for the pardonable optimism naturally
to be expected on the part of the manufacturer.
The accompanying table published in the "Gas Solicitor's
Handbook" was believed at the time of publication to be repre-
sentative of the results that might be expected in actual service
under good conditions with clean lamps and new mantles of
good quality, and laboratory tests have indicated that the de-
preciation which results from burning over the reasonable period
of time elapsing between maintenance calls should be negligible.
philbrick: store lighting 501
It was desired to substantiate this in practise, and 1 was very
much gratified to find that this was done.
Lacking facilities for determining the gas consumption in each
case 2>TA cubic feet per hour was taken as presumably a close
approximation to the actual consumption, the sizes of the mantles
indicating that on gas of the quality furnished, this assumption
was reasonably accurate.
Tests made upon new mantles of known efficiency confirmed
this opinion.
In determining upon a procedure for these tests, horizontal
illumination upon the working plane was selected as the basis.
Not because this is the only plane requiring illumination but
because with practically all the glassware usually sold for store
illumination, a sufficient degree of horizontal illumination is
always accompanied by at least a sufficiency of illumination
upon other planes. The main purpose was to obtain information
which might assist us in furnishing illumination to our customers
under the most favorable conditions.
There are several matters that are worthy of some special
attention in connection with these different installations. It will
be noticed that in many cases the illumination was measured at
comparatively few points, and while the numerical average of
the results would not give the actual average obtained throughout
the entire area, the numerical average is in all probability some-
what below the true average which would have been obtained
had a greater number of readings been taken. For instance, in
both Figs. 1 and 2, it will be seen that most of the measurements
were taken in the more poorly illuminated portions of the room,
very few being taken at points immediately beneath the lamps
where the illumination would be the highest. This statement also
applies to Figs. 3, 4 and 5. With regard to Fig. 7 the actual
average illumination is probably far above the figures which we
have given, possibly as much as 20 to 30 per cent., on account
of the fact that the lamps were quite low and most of the read-
ings were taken at situations so far from the lamps that they
were outside of the range of effective distribution. In this par-
ticular instance, of course, the really important consideration is
the amount of light on the face of the customer, and the lamps
9
502
TRANSACTIONS I. E. S. — PART II
are properly arranged to distribute the light for this purpose in
the best manner, and the general illumination of the room is
of much less importance.
/
6
If »:
Is !
t
.-If.
.J
*
g
|s
Hh\/
o _
In the store shown in Figs. 7 and 8, it will be noticed that a
greater number of readings were taken and these probably rep-
resent truer averages than any of the others.
An interesting point is that, according to the Gas Solicitor's
Handbook, (page 30), is claimed that a cubic foot of gas per
philbrick: store lighting
503
hour will produce about no effective lumens in a room with
light side walls (that is, a sufficient amount of light to illuminate
'Uiili-ftif
Fig. 3. — Plan of shoe store.
a
1
1
r«
«E
a:
u
t
z
0
0
COUNTER
05
a
04
o3
-8
Ot
ov
n
■t
Ol
a
SCALE OF FT
Fig. 4. — Plan of jewelry store.
SCHLC OF TUT
Fig. 5. — Plan of pool room.
0*
*
OT
»•
■ , .
9
m
-
..
0*
„
1 10 square feet to an intensity of 1 foot-candle) and with dark
walls 100 effective lumens. From the numerical averages, as ob-
tained in the test, it will be noted in the last column that there
504
TRANSACTIONS I. E. S. — PART II
was only one gas installation which dropped a sufficient degree
below the nominal efficiency stated in the handbook to make the
matter worth considering, and this is the barber shop, in which
as I stated above, the lamps were so low that most of the readings
came outside of the effectively lighted area, so that as far as in-
dicating the efficiency of the gas service and maintenance, this
should be eliminated.
The installation shown in Fig. I had not been maintained for
two months, and was only i per cent, below the nominal efficiency.
Installation No. 2 was 3 per cent, below.
□
CISAR
COUNTER
ml
on 0
«»i q °*
m
A
2T X X a
I I I I 1-
iCAit in rtrr.
Fig. 6. — Plan of barber shop.
The value of tests of this character, both to the gas company
and to the industry at large, is to my mind quite apparent and
extends not only to the solicitation of new business, but to the
proper design of lighting systems and to the maintenance of
existing installations.
As regards the soliciting of new business, actual tests of in-
stallations of which the prospect has personal knowledge rein-
forces the arguments of the salesman most forcibly.
In many cases, contracts for lighting hinge mainly upon econo-
PHILBRICK : STORE LIGHTING
505
mic considerations. Every salesman claims the highest economy
for the particular illuminant he happens to be selling. His claims
are of necessity based upon laboratory tests, or upon service tests
in other localities in which the conditions may or may not ap-
proach those in his own situation. He can submit no evidence
of the validity of his claims that has much weight with the cus-
tomer. In such cases tests upon installations in the same locality
are very convincing, particularly if some of them happen to have
been made upon the premises of the customer.
/
wrwv c
~/
/■■
w
I I I T~T
I' I" f "I 1
fl
Fig. 7. — Plan of shoe store.
£
I ' !
TTTCL!; M-'=
Fig. 8.— Plan of carpet and rug store.
In the case of the barber shop quoted above, the proprietor
himself took simultaneous readings with the operator and saw
for himself the true comparison between competitive illuminants
so far as the amount of light he was getting was concerned, and
even though the amount of light fell 43 per cent, below the theo-
retical, the illumination was nearly twice that which was given
when the shop was lighted by the competitive illuminant.
A test like this gives the consumer an exact idea of the com-
parative value of competitive illuminants and makes a valuable
reference for the solicitor of the gas company, inasmuch as the
conditions are approximately the same under any city and the
solicitor has but to refer to the test and the consumer personally
for a confirmation of the data with which he is trying to secure
the business of a prospective customer.
506
TRANSACTIONS I. E. S. — PART II
Photometric Data.
Fig. i.
Fig. 2.
Station
Foot-candles
Station
I
1.6
I
2
2.6
2
3
3-8
3
4
4.0
4
5
4-3
5
6
4.1
6
7
3-7
7
8
2-7
8
9
2-3
9
IO
5-2
10
II
7.8
11
12
6.8
12
13
7.0
*3
14
6.7
14
15
6.0
15
16
3-65
16
17
1.6
17
18
3-2
18
19
4-5
19
20
4-5
20
21
4.6
22
4-5
23
4-9
24
2.4
Foot-candles
6.0
5-3
4-5
3-3
4-5
5-2
4.2
1.85
5-8
5-3
3-7
1.7S
5-i
5-o
4.0
2.2
5-6
3-3
1.74
2.2
Fig- 3-
Fig. 4-
Station
Foot-candles
I
3-1
2
3-9
3
5-7
4
7-3
5
4.8
6
6.6
7
2.25
8
5-8
9
5-i
10
2.0
11
6.4
12
4.0
13
6-3 -
14
4-4
15
3-2
16
3-°
Station
Foot-candles
I
2.2
2
5-5
3
7-8
4
7.0
5
4.0
6
i-9
7
4.25
8
8.0
9
7.2
10
4-7
PHILBRICK : STORE LIGHTING
507
Fig. 5-
Fig. b.
Station
I
Foot-candles
I.30
Station
I
Foot-can lies
3-6
2
I. OO
2
6.2
3
4
I.l6
.80
3
4
3-5
2-5
5
6
.90
13.00
5
6
5-6
3-o
7
6.50
7
4.0
8
12.50
8
7-3
9
5-50
9
i-5
10
9.20
10
4-7
11
12
I.46
.86
11
12
7.0
2.8
13
1.77
13
3-3
14
*-i3
14
4.4
15
1.70
15
1-45
Fi
Station
I
g- 7-
Foot-candles
2.20
Station
I
Fig. 8.
Foot-candles
i-34
2
3.00
2
2.09
3
4-5o
3
2-74
4
5
6
5.00
3.60
2.60
4
5
6
2.40
LIS
1-34
7
4-3o
7
1.96
8
5-15
8
309
9
4- 25
9
3-05
10
3.00
10
1-52
11
2.30
11
1.70
12
13
14
3.60
6.08
5.26
12
13
14
3-°3
3-53
3-47
15
16
3-54
3.20
15
1.47
17
18
4.45
♦ 5-4o
19
4.20
20
3.10
21
2.70
22
3.00
23
3.65
24
3-75
25
26
2.90
2.20
•
27
2S
3-55
4.10
29
3-55
SO
2.40
5o8
TRANSACTIONS I. t. S. — PART II
Gas
Fig.
No.
I
2
3
4
5
6
7
Fig.
No.
I
2
3
4
5
6
7
8
Area Ceiling
Business sq. ft. height
Tailor shop 979 i4/-o//
Lunch room 2,704 \2'-&f/
Shoe store 2,066 i2/-o//
Jewelry store 450 io/-6//
Poolroom 1,487 y'S"
Barber shop 611 8'-&>
Shoe store 1,916 i5/-o//
Carpet store 2,31s i8/-o//
Lamps per
Business outlet
Tailor shop 2
Lunch room 14-2L., 3-1L
Shoe store 2-1L., 1-2L.,
6-3L., 2-4L.
Jewelry store 2
Pool room 4
Barber shop 2
Shoe store 2
Carpet store 3
Height of
lights
9'-
Walls
Medium
No. of
outlets
6
io'-6"
Medium
17
$'-6"
Light
11
&-&'
Dark
4
7'-8"
Dark
5
y/_2//
io'-o"
Light
Light
6
10
il'-o"
Dark
6
Gas
Type
6 Reflex
Reflector
Extensive prismatic
Fig.
No.
I
2
3
4
5
6
7
Total con.
Business per hour
Tailor shop 40.0 Cu. ft.
Lunchroom 102.3 "
Shoe store 99.0 "
Jewelry store 26.4 "
Pool room 66.0 "
Barbershop 39.6 "
Shoe store 66.0 "
Carpet store 59.0 "
Illumination
Average Minimum Maximum
Eff. I,m. per
cu. ft. or watt
Fig. 1 ' .
No. Business Actual Theor.
i Tailor shop 104.0 105
2 Lunch room 108.0 105
3 Shoe store 96.2 no
4 Jewelry store 90.0 100
5 Pool room 88.3 100
6 Barbershop 62.6 no
7 Shoe store 109.5 no
8 Carpet store 87.6 100
* Increase
4.26
T
6
7-8
4.2
I
74
6.0
4.61
2.0
7-3
5-25
I
■9
8.0
3-92
O
8
13.0
4.06
I
45
7-3
3-76
2
.20
6.08
2.25
I
15
3-53
Gas
Per cent.
Actual
below
Time since
theor.
maintained
O.I
2 Months
*3-Q
I Day
12-5
Not maintained
10.0
1 Month
12.0
1 Month
43-o
1 Week
0.5
1 Week
12.2
1 Month
PHILBRICK : STORE LIGHTING 509
I believe that it would be of much value to the commercial de-
partments of all the gas companies to have tests like the above
made and printed and copies given to their solicitors, thus enab-
ling them to meet the question of the consumer "How do you
know" with data which applies to the question at hand and not
an irrelevant mass of figures which mean nothing under the
local conditions.
The great feature of help to solicitors is not that the consumer
will understand terms of illumination intensity but that he can
see for himself the real value of his lighting and the effect on his
pocket-book in dollars and cents which to him is a thing of vital
interest.
Heretofore the designs of installations have been made at ran-
dom, the lighting effects have often been not satisfactory upon
the first trial, a second and sometimes the third attempt being
necessary to give the consumer the proper light in the proper
place. This often entails considerable expense to the consumer
and company and creates dissatisfaction.
It is a fact that the lighting engineer of the manufacturer has
been at the disposal of the gas companies for sometime back,
offering to lay out installations and perform all the illuminating
engineering work, but how many of the companies have taken ad-
vantage of this offer. If solicitors would make a series of tests,
and compute a table of lighting efficiency in various store-rooms
they could easily prevent mistakes in future installations.
Manufacturers' hand-books are all right, but they are too
often discounted or ignored, but data gotten right in the field
under the local conditions cannot be discounted or ignored but
are convincing facts.
There is one test which I did not have time to prepare but
would have been advisable to undertake and that is a test on
temperature at various points in our stores under the gas and
electric lighting, taking into account the outside temperature dif-
ference on the different days of test. I believe the heat objection
against gas lighting can be greatly diminished, if not eliminated.
Lastly, it is realized that the efforts of the illuminating engineer
are entirely lost if the lamps are not kept clean and the mantles
renewed at proper intervals. It is absolutely necessary to have
5IO TRANSACTIONS I. E. S. — PART II
men who are careful and intelligent, who have some knowledge
of the lamps they are cleaning; and who know when the adjust-
ment of the lamp after cleaning is as near right as possible to get
maximum efficiency. Too often do companies try to economize
on this most important part of their business by employing boys
at very small fixed wages, and expecting them to have the incli-
nation to become illuminating experts. To my mind the only
way to pay lamp maintainers is on the sliding scale plan with
the deduction for complaints on lamps which they have maintained
and caused the consumer trouble. The lamp, whether gas or
electricity, must have clean glassware and bulbs for much business
is lost by our failure to attend to these features. Of course, it
costs more to maintain lamps properly than it does to clean them
in a half-hearted manner, trying to keep the cost down at the
expense of the illumination. This, however, can be met by a
proper and not excessive charge based on the sliding scale of
consumption per mantle. (It is obviously not fair that the large
consumer of gas or other illuminants should pay at the same rate
for his maintenance as the consumer who uses his lights merely
as a makeshift.) Below is the maintenance schedule which is in
use by a gas company in York, Pa., which was adopted after care-
ful consideration by the officers of the company. This schedule is
giving entire satisfaction to the consumer and company alike.
YORK GAS COMPANY
York, Pa., 1913.
The undersigned at number Street,
York, Pa., hereby makes application to the York Gas Company to use gas
for illumination, at the regular rate of the Company.
It is further understood that this application, being approved, the York
Gas Company agrees to loan and install the necessary Gas Lighting Fixtures,
Piping, etc. , free of cost.
In all cases where combination Gas and Electric Fixtures are specified
and installed, the undersigned agrees to use Gas regularly on such fixtures
from September 1st to May 1st, that said fixtures remain in his premises.
These fixtures will be given regular monthly inspection, new mantles sup-
plied as needed, be cleaned and adjusted and be given such other additional
attention as may be necessary to keep fixtures in good working order.
In consideration of the above the undersigned agrees to pay maintenance
according to the following scale:
PHIEBRICK : STORE LIGHTING
5"
Consumption per mantle per month,
(< « <
50 feet or less
100 '
< 11 <<
125 '
1 II II
150 '
1 <1 II
175 '
• II <<
200 '
I II II
225 '
' " "
250 '
I 11 II
275 '
I II II
3°° '
1 II II
400 '
1 11 11
500 '
1 II II
600 '
1 II II
700 '
I II 11
800 '
I II II
1,000 '
1 II 11
1,250 '
1 II II
1,500 '
III II
2,000 '
1 II II
•15
cents
per month
•IO
"
"
"
■ 9*2
"
11
"
• 9
"
"
"
• sy2
"
"
it
. 8
1 1
11
"
• 7H
"
"
1 1
■ 7
"
"
"
■ 6/2
"
"
"
. 6
"
"
it
• 5%
11
"
1 1
• 5
"
"
1 1
• AlA
"
"
"
• 4
<i
"
•'
• 3*A
"
' '
"
• 3
it
"
"
• 2/2
"
"
"
• 2
"
1 1
"
• • I
11
11
All maintenance to be paid monthly.
Orders taken by Signed ■
Finally, gas men should appreciate the possibilities of gas light-
ing, determine exactly the intensity of light in store-rooms,
factories, etc., approach consumers with more intelligible, con-
vincing data, and keep the lamps clean and properly adjusted.
By doing this, it has been said, "Two units can be made to
grow where only one grew before."
The writer wishes to acknowledge his indebtedness to Mr. R.
F. Pierce, for his help in the preparation of this paper.
DISCUSSION.
Mr. R. F. Pierce : Inasmuch as these statistics were taken
for a purely commercial purpose and intended to set forth only
the existing lighting service conditions in actual installations, no
deductions into which enter such considerations of gas pressure
or electric voltage should be drawn. While these data were
taken, they were not presented in the paper which is simply a
statement of lighting conditions in a certain locality under the
conditions found, and having no reference whatever to any other
set of conditions.
I think Mr. Philbrick rather over-estimates the amount of as-
sistance that he received from me. I furnished the illumination
measurements, and in that connection should like to explain the
512 TRANSACTIONS I. E. S. — PART II
method used. The measurements were comparatively few, the
idea being to obtain some sort of an average figure which should
at least be low. It was impossible to devote the time necessary
to take a larger number of measurements which would have
given more accurate figures, but the results were comparative
rather than absolute. Mr. Philbrick wished particularly to ob-
tain an idea concerning his position in the competitive field and
to ascertain how closely the published figures for effective lumens
per cubic foot were approached in his situation.
In this connection, I would emphasize that the table given on
the first page of the paper does not refer to measurements ob-
tained at York, but to figures for effective lumens per cubic foot
published by a manufacturer of gas lighting appliances for use
in the design of installations. These figures represent averages
of a number of practical installations and have been given out
as a practical basis for the design of gas lighting systems.
In the tests at York a number of installations were selected
in which combination fixtures were used with the same type of
reflector on both gas and electric outlets and measurements were
made on both systems at the same points to show a comparison
between the two. The gas company had no prior knowledge
either of the date of the test or the installations that would be
tested. The latter were picked out at random without reference
to the condition of the lamps or of the installation. The tests on
the gas lighting installation showed a maximum discrepancy be-
tween the data published by the manufacturer and the results
obtained under these tests of about 12.5 per cent., with the single
exception of one installation in which the number of test sta-
tions was so small that, on account of the very low height of the
lamps, the majority of the test stations fell outside of the effective
radius of the lamps. In this installation the discrepancy between
published figures and the average obtained was 43 per cent.
That this discrepancy was due to the location of the stations is
indicated by the fact that a test of the electrical installation at
the same points showed a discrepancy of 60 per cent, from the
commonly used data for effective lumens per watt, so that in
both cases the discrepancy appears to be due to the condition
mentioned. I do not mean to convey the impression that the
relative service conditions found in this situation exists in the
STORE LIGHTING 513
majority of plants. As I remember the figures, the average dis-
crepancy for effective lumens per watt obtained in the electrical
installations was in the neighborhood of. 40 to 45 per cent. It
is quite obvious that this was due to a large extent to failure to
replace the lamps at proper intervals and to select lamps of
proper voltage. A comparison between the two services em-
phasizes the advisability of central stations and gas companies
obtaining actual illumination measurements in order to deter-
mine the lighting service really given to their customers under
the conditions of actual use. The electric service in this particu-
lar situation was plainly below the average, while the gas lighting
service was much better. There are, of course, many- cases in
which this comparison would be reversed. The results of this
test emphasize the fact that no general comparison between the
service rendered by the two illuminants is possible. The most
important factor is the quality of service rendered in each par-
ticular situation.
Mr. T. J. LitlE, Jr. : It is very agreeable to note in Mr.
Philbrick's paper that the data that he has obtained at York com-
pares very favorably with the published data on the same types
of lamps which he tested. It is a fact that published data on
many of these lamps is in many cases conservative from the
fact that the tests were made under 2.5 inches water pressure,
while in many cities the service pressure is much higher than
this, and they consequently get increased candle-power and effi-
ciency from their burners.
For instance, the consumer's service pressure in Chicago is
6 inches while in San Francisco it runs considerably higher than
this. It may be interesting for you to know that considerably
higher efficiencies than those which have already been published
are expected in the near future. They have already' been reached
experimentally. For instance, it is quite possible with certain
sizes of inverted lamps to exceed 30 candles per cubic foot,
mean lower hemispherical rating. In -fact, I have seen as high
as 53 candles per cubic foot, mean lower hemispherical rating,
the lamp burning on low pressure {2l/2 inches water pressure).
It is a remarkable fact when looking back over the develop-
ment of gas and electric lighting systems that in the year 1880 the
514 TRANSACTIONS I. E. S. — PART II
most radical developments took place, namely, the Edison in-
candescent electric lamp and the Welsbach incandescent gas
lamp. Ever since that date, whenever there was a radical de-
velopment in the one system, there was something brought out
in the other to meet it.
Mr. Ward Harrison : Mr. Philbrick's paper is very inter-
esting to me. It shows what an effective weapon poor service
on the part of one lighting company may afford its competitor.
It has been found, however, that comparative, tests of this
character too frequently lead to bad feeling between the lighting
companies interested. This feeling can be changed to one of
good natured rivalry simply by making it a practise in each case
to notify everyone interested in the test and arranging for them
to attend. It has been found, also, that when the prospective pur-
chaser and the representatives of all the competing interests are
present at the time of the tests, much of the dissatisfaction in
regard to test methods and results which so often follows, is
obviated.
Mr. E. B. Rowe : I cannot emphasize too strongly Mr. Har-
rison's contention that operating companies should know what
results the customers are getting from service on their lines. The
making of installation tests similar to those covered in Mr.
Philbrick's paper is certainly to be commended, but the condi-
tions under which test is made should be clearly recognized. In
making use of such test results complete data should be given
or the results should be reduced to a common equivalent basis.
For instance, on the third page no reference is made to a
measurement of the gas pressure, which of course is one of the
variables in the test and has a considerable influence on the re-
sults obtained. In the same way in tests on an installation of
electric lamps the exact voltage on which lamps were operated
during the test should be stated and the effect of voltage varia-
tions made clear to the layman or the results could be reduced
to the normal voltage or gas pressure. Personally I have some
hesitancy in making use of published data unless all the facts
are at hand and I believe care should be taken in presenting test
results to cover all questions which are likely to arise in the
minds of those who may wish to use the data.
LAW AND POWELL: STORE LIGHTING 515
DISTINCTIVE STORE LIGHTING.*
BY CLARENCE L. LAW AND A. L. POWELL.
Synopsis: Certain stores, namely the high-class shops, demand a
striking individuality of design. This paper describes in detail several
typical lighting installations which have come to the attention of the
authors; it discusses a particular store of each of several classes: shoe,
millinery, toy, candy stores, etc. Data are given as to the dimensions,
wall, ceiling and floor coverings, arrangement of fittings, type of glass-
ware, number and sizes of lamps, and a general description of the appear-
ance of each store. This information may not be directly applicable to
the design of a new installation, but since the lighting systems outlined
are giving satisfactory service, the quantitative element may be of use
and the novelty of some of the equipments may suggest to the designing
engineer ideas which will be applicable to his particular problem.
If one pauses to analyze retail places of business as a whole,
it becomes evident that a convenient and complete classification
may be made as follows : ordinary small stores, large dry goods
and department stores, and high grade shops.
The authors treated the first of these classes in a paper pre-
sented at the last convention1 of the Society; they realized that
high efficiency of light utilization, low initial cost of installation,
low maintenance and simplicity were the determining factors. It
seemed desirable to suggest a standard practise, and an attempt
was made to do this on the basis of averages obtained from an in-
vestigation of a large number of stores of this class. Some
criticism was elicited to the effect that this scheme would pro-
duce a monotonous condition, but the fact still remains that
artistic appearance cannot be had cheaply, and, due to the small
profit earned by a store of this sort, the amount spent for light-
ing is, of necessity, small.
In the large store, efficiency and artistic appearance become
more nearly balanced. Artistic lighting implies good diffusion ;
and with the present commercial illumi'nants, this cannot be had
* A paper read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
1 Transactions of the I. E. S., Vol. VII, p. .
5l6 TRANSACTIONS I. E. S. — PART II
without some absorption of light. The merchant can afford to
spend a relatively larger sum for lighting than could be spent
by the small storekeeper, and some sacrifice of light is made to
obtain better diffusion. The store should have a harmonious
system of lighting for the main parts of the entire building, yet
there are some parts which are, in reality, shops, and should be
so treated. The general requirements for department store
lighting have been discussed several times in the Transactions*
of the Society, and there is no need for their repetition in this
paper.
A high grade shop should be considered quite differently from
the monotonous store which is so common in large cities. Un-
like a small store, it should be considered individually and with
respect to its particular line of business. This shop is, as a rule,
small, handsomely and lavishly furnished, splendidly finished
to the minutest detail, and located in the most fashionable sec-
tion; it handles only the best grade of goods (frequently im-
ported) and sells to a discriminating class of customers. The
proprietor or manager is willing to spend large sums for the
right equipment and maintenance. The profits for each indi-
vidual piece of merchandise sold are undoubtedly greater than in
other stores, and therefore, more money can be spent for indi-
viduality of equipment. Artistic appearance is the predominant
factor, and, therefore, a distinctive system of lighting is neces-
sary, efficiency of the installation being a secondary consideration.
It should be the aim and desire of shopkeepers of this class
to interest and attract prospective customers, making them per-
manent habitues of their stores. Some definite architectural
scheme should be carried out or symbolism expressed. Many
stores show the influence of the personality of the proprietor, and
often such details as the dress of the sales force are in harmony
with a certain predetermined plan.
Among the points which should be given consideration by a
shop proprietor in planning a distinctive store, may be mentioned
the following:
Design of the exterior; woodwork of the interior; color of
* C I,. Law and A. J. Marshall, "The Lighting of a Large Store," Vol. VI (1911),
p. 186; H. W. Shalling.— "Department Store Lighting," Vol. VIII (1913), p. 17.
LAW AND POWELL: STORE LIGHTING 51/
walls and ceiling; finish of show cases; floor covering; finish
and type of lighting fixtures, glassware and lamps.
Numerous examples of distinctive store lighting have un-
doubtedly come to the attention of every one, but a description
of a few which the authors have observed may indirectly suggest
schemes which will prove of benefit and aid in the advancement
of the art of lighting. In last year's paper, the quantitative
element in designing the lighting was discussed. A number of
stores were grouped under one heading, but with the case in
hand where individuality of stores of any one type is the essen-
tial, it is necessary to use the "case" system, illustrating by
example.
TOY STORE.
An effective lighting system of a toy store may be seen in F.
A. O. Schwartz's store on Fifth Avenue, New York City. (Fig.
1). The building is of modern construction with a high ceiling
supported by pillars. The entire interior, including walls, ceil-
ing and show cases, is finished in white, affording an excellent
background for the varicolored toys on exhibition. The store
was formerly lighted by Nernst lamps with massive ornamental
housings finished in gilt. These were of Renaissance design,
and in keeping with the capitals of the columns in the main room.
The heart-shaped Nernst globe was replaced by a 14 in. (35.56
cm.) opalescent glass acorn type diffuser, and 400-watt clear
tungsten lamps were used. Diffusion is good and shadows from
pillars and overhanging shelves are minimized.
Store Window
Length 145 ft. (44.20m.). 60-watt bowl frosted tungsten
Width 41 ft. (12.50 m.). lamps set in recessed mirrored
Approximate area 6,000 square pockets at front edge of window,
feet (557.42 sq. m.). spaced about 3 ft. (0.914 m.).
Ceiling height 17 ft. (5.18 m.).
Lamps 12 ft. (3.66 m.) from floor.
15 400-watt, 4 250-watt clear
tungsten lamps.
Total wattage 7,000.
Watts per sq. ft. (0.30 sq. m. ) 1.2.
JEWELRY STORE.
Richness and splendor are symbolized by jewels and, therefore,
the shop dealing in these, should be magnificently finished. The
10
518 TRANSACTIONS I. E. S. — PART II
store of E. M. Gattle, on Fifth Avenue, New York, Fig. 2 can
well be used as an illustration. All of the show cases and fur-
niture are of mahogany; immense gray marble columns and
pilasters with gold capitals support a paneled ceiling, which is
also of mahogany finish. The parts of the side walls not oc-
cupied by window space are a green tint, decorated in gold.
The floor is of oak in parquet style. Light is furnished by
eighteen shower fixtures, verde finish, using 40-watt all frosted
round bulb tungsten-filament lamps, and in the paneled recesses
in the front part of the store are eight cut glass hemispheres,
accommodating two 25-watt clear tungsten lamps each. A high
wattage is necessary with this system in a room of this style, as
the reflection coefficients of the ceiling and walls are very low.
Store Window
Length 78 ft. (23.77 m.). 2 aluminum finish trough refiec-
Width 38 ft. (11.58m.). tors.
Area 2,960 sq. ft. (274.99 sq. m.). 25-watt tungsten lamps spaced
Ceiling height 13 ft. (3.96 m.). about 10 in. (25.4 cm.).
Lamps 10 ft. (3.04 m.) from floor. Backing of window green plush.
220 40-watt round bulb all
frosted 16 25-watt clear tungsten
lamps.
Total watts 9,200.
Watts per square foot 3.1.
TOGGERY OR HABERDASHERY SHOP.
As this class of store caters entirely to men, the store fittings
should not be radical to any appreciable extent. Neatness, sim-
plicity, and up-to-date appearance should characterize the shop.
The lighting system must be quite efficient, as a high intensity
of illumination is desirable.
Fig. 3 shows a night view of the installation of one of the
shops of Weber & Heilbroner on Broadway, New York City,
which conforms excellently with the above requirements. Six-
arm brush brass fixtures of well balanced proportions are used
with clear 100-watt tungsten lamps and opalescent bowl shaped
reflectors. Show cases, counters and woodwork are of polished
mahogany; ceiling smooth white plaster; walls above shelves
covered with green burlap, and floor of hard wood. The window
trim is of Circassian walnut, forming an excellent contrast to the
-
6')*
Fig. i. — Distinctive illumination of a toy store.
Fig. 2. — Distinctive illumination of a jewelry store.
Fig-. 3. — Distinctive illumination of a haberdasher's store.
Fig. 4. — Distinctive illumination of a millinery store.
LAW AND POWELL: STORE) LIGHTING 519
dark blue velvet backing for the goods on display. A white fixed
shade, extending to within six feet (1.83 m.) of the sidewalk
level, serves as a valance.
Store Window
Length 69 ft. (21.03 rn.). 100-watt clear tungsten lamps.
Width 18 ft. (5.49 m.). Concentrating prismatic reflector.
Area 1,342 sq. ft. (124.77 sq. m.). Spaced 14 in. in a row along center
Ceiling height 13 ft. (3.96 m.). of false ceiling.
Lamps 10 ft. 6 in. (3.20 m.) from
floor.
30 100-watt clear tungsten lamps.
Total watts 3,000.
Watts per square foot 2.4.
MILLINERY.
Since Paris is the seat of fashions, to create the proper atmos-
phere, the display room should be "Frenchy" in character.
Mme. Bruck's shop on West Fortieth Street, New York City,
shown in Fig. 4 may be taken as an example. White show
cases, covered with mirrors line the walls, and the dainty furni-
ture is all finished in white enamel. White has the advantage
that it does not "clash" with the colored materials of the hats and
tend to divert the attention from the goods on display. The ceil-
ing is of smooth, white plaster and a border of satin finish wall
paper matches the old rose Wilton carpet and silk window hang-
ings. Two ten-light brass finish shower fixtures with bowl-
frosted tungsten lamps surrounded with crystal beaded glass,
furnish general illumination. Localized illumination at the
mirrors is supplied by side wall brackets, brush brass finish, Em-
pire style, equipped with bowl-frosted tungsten lamps, shielded
by crystal and old rose beaded shades. A few plants add to
the attractiveness of the room.
Store Window
Length 40 ft. (12.19 m.). 25-watt clear tungsten lamps in
Width 12 ft. (3.66 m.). concentrating prismatic reflectors
Area 480 sq. ft. (44.59 sq. m.). on 2 ft. centers.
Ceiling height 10 ft. (3.04 m. ). 3 25-watt tungsten lamps in crystal
Lamps 9 ft. (2.74 m.) from floor. fixtures in center of window.
Total watts 750. 2 side wall brackets, cut glass shade
30 25-watt bowl frosted tungsten and 25-watt tungsten lamps.
lamps.
Watts per square foot 1.5.
520 TRANSACTIONS I. E. S. — PART II
CANDY STORES.
One of the newest and most attractive of New York's Fifth
Avenue stores is that of Schrafft, a view of which is shown in
Fig. 5. A combination of semi-indirect and totally indirect
illumination is used. The front portion of the store serves as a
shop and is lighted ,by five three-light carved alabaster bowls
suspended from the ceiling by silk-covered supports, and four
one-light bowls, two on brackets and two on short pillars. The
ceiling here is tan decorated with raised gold figuring ; walls are
elaborately decorated, with red, green and blue on a neutral
backing. Show cases are of Circassian walnut; pillars and floor
of marble. A number of small decorative standards are used
to illuminate the counters. The rear half of the store is used as
a lunch room. In the center of this room is what is apparently
a fern-covered urn. This contains a white enamelled reflector
and a cluster of clear lamps, the light from which is directed to
the cream colored ceiling and walls, lighting the room indirectly.
Store Window
Length 74 ft. (22.55 m.). Finished in Circassian walnut; roof
Width (average) 17 ft. (5.18 m.). recessed with mirrored pyramidal
Area 1,200 sq. ft. (111. 48 sq. m.). reflectors and 25-watt clear tung-
Ceiling height 14 ft. (4.27 m.). sten lamps installed in squares on
Lamps 10 ft. (3.05 m.). 18 in. centers.
5 150- watt clear tungsten lamps.
4 40-watt clear tungsten lamps.
15 60-watt clear tungsten lamps.
Total watts 1,810.
Watts per sq. ft. 1.5.
Delicious sweets of great variety originate in the Far East,
and an Oriental scheme of decoration for a candy store is, there-
fore, often appropriate. Fig. 6 shows a night view of the ex-
terior of Page & Shaw's Fifth Avenue Shop. It can be seen that
the window is partially covered with a delicate tracery of red,
green and blue leaded glass ; at night this is accentuated by illum-
ination from lamps in the ceiling of the window. Three metal
and art glass hanging fixtures are also part of the window equip-
ment.
Free use of the primary colors is made in the decorating of the
walls and ceilings of the store with conventional Moorish figures.
The floor is of composition, red and white mosaic. It can be
safely said that no two of the interior lighting units are alike:
Oriental metal and colored glass domes, pottery vases lighted
LAW AND POWELL: STORE LIGHTING 521
from within and silk-covered lanterns furnish a very low inten-
sity of general illumination, with a higher value on the counters
and show cases.
The cashier's desk is surrounded by leaded glass made in the
form of a miniature Turkish house, the whole surface of which
is illuminated by a number of line source tubular tungsten-fila-
ment lamps concealed in its interior.
Store Window
Length 30 ft. (9.14 m.). Roof recessed with mirrored pyra-
Average width 15 ft. (4.57 in.)- midal reflectors ; one ft. centers ;
Area 450 sq. ft. (42.8 sq. m.). clear 16 c-p. round bulb carbon
Ceiling height 10 ft. (3.05 m.). lamps; 2 60-watt all frosted
Lamps 5 to 7 ft. (1.52 to 2.13 m.) round bulb and 3 25-watt regu-
from floor. lar tungsten lamps in hanging
Total watts 900. lanterns.
Watts per square foot 2.0.
GROCERY STORE.
A neat, attractive display will cause trade to flock to the store
which is properly arranged. Cleanliness is a very important
point to remember. There is no demand for a system of dec-
oration for this class of store, but the walls, pillars and ceiling
should have frequent painting. A dark wainscoting, the color
of the shelves and show cases, with neutral walls and ceiling,
makes an attractive combination. Almost any lighting unit which
is neat and inconspicuous will serve.
D. M. Welch & Son's store in New Haven, Conn., shown in
Fig. 7 serves as an illustration of the above requirements. The
counters and show cases are of hard wood, natural finish ; the
trim is dark green ; and the ceiling and walls are painted a light
tint. Neatness is particularly characteristic. Two hundred
and fifty watt tungsten lamps, in totally enclosing prismatic re-
flectors are used for general illumination. The unit is efficient
and a satin finished lower half provides excellent diffusion. A
short brush brass chain with canopy serves as the fixture.
Store Window
Length 70 ft. (21.33 m.). 100-watt clear tungsten lamps in
Width 40 ft. (12.19 m.). concentrating prismatic reflec-
Area 2,800 sq.ft. (232.25sq.m.). tors spaced 2 ft. 6 in. (0.76 m.).
Ceiling height 14 ft. (4.27 m.).
Lamps 11 ft. (3.25 m.) from
floor.
10 250-watt tungsten lamps.
Total watts 2,500.
Watts per square foot 0.9.
522 TRANSACTIONS I. E. S. — PART II
TEA ROOM.
Coziness is the keynote of success of these establishments. A
number of years ago a young woman started in a small way to
sell home-made candy and pastry among her friends. Her
energies soon developed into a methodical business system, and
her products sprang rapidly into favor, with the result that
"Mary Elizabeth" has branch shops in many of the large cities.
Her New York store, which is shown in Fig. 8 and located on
Fifth Avenue, is finished in white on the outside, with her fac-
simile signature in black serving as a sign.
The shop itself is modelled after a New England interior of
fifty years ago; the ceiling is low and finished in white plaster;
the floor of wide boards is painted a dark yellow and covered
here and there with rag carpet "runners". The tea room proper
is in the rear. On the right is an old fashioned fire place, and
on the left a number of "stalls" similar to those found in taverns
of bygone days. Small tables, covered with spotless linen, and
gilt chairs are arranged as shown in the illustration. Shelves,
counters and windows are trimmed with dainty white material.
Light is furnished by tungsten lamps in shirred silk shades
which have a slight touch of color. Sixteen of these are at-
tached to ceiling outlets and eight are on wall brackets.
The atmosphere of the room is extremely inviting and the
scheme of decoration well executed.
Store Window
Length 62 ft. (18.90 m.). 40-watt tungsten lamps in shades
Width (average) 18 ft. (5.49m.). as used in the store; row in the
Area 1,110 sq. ft. (103. 11 sq. m.). center of ceiling; 2 ft. (0.61 m.)
Ceiling height 8 ft. (2.44 m.). centers.
Lamps 7 ft. (2.13 m.) from
floor.
24 40-watt clear tungsten lamps.
Total watts 960.
Watts per square foot 0.9.
SHOE STORE.
Most stores of this class have a center bench arrangement, the
entire wall space being covered with boxes on shelves. A room
of medium width will require at least two rows of units to give
satisfactory illumination on the labels on the boxes and at the
foot rests where the shoes are fitted and inspected.
A particularly novel layout is shown in Fig. 9, a night view
f>'
Fig. 5. — Distinctive illumination of a candy store.
Fig. 6. — Distinctive illumination of a candy store window.
Fig. 7. — Distinctive illumination of a grocery store.
Fig. 8.— Distinctive illumination of a tea room.
LAW AND POWELL: STORE LIGHTING 523
of Frank Brothers' Fifth Avenue (New York) shop. Entering
from the street, one passes into the rotunda (shown in the back-
ground of the photograph) about 16 feet in diameter, the dome
of which is supported by Corinthian columns. The floor is of
mosaic marble and the ceilings, cream colored, with raised plaster
decorations. Show cases, with attractive dressings, are grouped
about the room. Suspended from the center of the dome is an
ornamental inverted fixture containing eighteen lamps. This
consists of six diffusing glass globes, pressed into the form of
huge shells; below these are four round bulb carbon lamps en-
closed in amber beaded glass.
The store proper is rectangular in shape and a balcony 6 feet
(1.83 in.) wide extends completely around the interior. The
cream colored ceiling beneath the balcony is divided by beams
into squares. In the center of each square is a lighting fixture
consisting of five pieces of pearl-like glass in the form of a large
shell ; a 40-watt clear tungsten lamp is located above each shell.
At the base of the shell is a 25-watt round bulb all frosted
tungsten lamp.
On each pillar from the balcony to the ceiling are located two
two-arm brass brackets with clear gem lamps in roughed
glass spheres. These serve to light the balcony and the center
portion of the store proper.
The oak parquet floor is partly covered with rugs : the furni-
ture is leather covered and the showcases and shelves are of
mahogany.
store The value of watts per square
Length 64 ft. (19.5 m.). foot would be of little sig.
Width 24 ft. (7.31 m.). nicance, as two types of lamps
Area (main floor) 1,540 sq. ft. are in use> and also both the
(14307 sq. m.). balcony and main floor are
Height under balcony 8 ft. lighted
2.44 m.).
Height above balcony 10 ft.
(3-05 m.).
17 25-watt round bulb tungsten
lamPS- Windows
85 40-watt clear tungsten lamps. Mirrored trough reflector with
64 50-watt clear Gem lamps. 50-watt Gem lamps outlets on
Total watts 7,000. 9 in. centers.
524 TRANSACTIONS I. %. S. PART II
ANTIQUE AND CURIO SHOP.
In many cases the lighting requirements of an antique store are
similar to those for a high class furniture store, that is, a low in-
tensity of diffused light suffices. Exposed light sources are very
obj.ectional, as the polished surfaces show the reflection and glare
is to be deplored in viewing the rare pieces on exhibition. Quite
often the lighting units themselves are "objects d'art." Such is
the case in the shop of Lewis & Simmons, shown in Fig. 10,
where hand carved alabaster bowls with clusters of clear lamps
furnish semi-indirect illumination.
The white ceiling, walls covered with dark red velvet and tan
velvet carpet, make a good color combination for displaying the
goods by contrast.
Store Window
Length 48 ft. (14.63 m.). Mirrored trough reflector with 25-
Width (average) 13 ft (3.96 m). watt clear tungsten lamps spaced
Area 630 sq. ft. (58.57 sq. m.). 8 in. (0.20 m.) apart along the
Ceiling height 14 ft. (4.27 m.). top of windows, and upright at
Lamps 8 ft. (2.44 m.) from the two sides to a height of
floor. about 4 ft. (1.22 m.).
20 40-watt clear tungsten lamps.
Total watts 800.
Watts per square foot 1.3.
BOOK STORE.
Scribners' new store, Fifth Avenue, New York, is an excellent
demonstration of a carefully planned and well-executed scheme
of lighting. The illustration, Fig. 11, shows very well the general
appearance of the room. The ceiling of the main bay is vaulted
and is of light gray sandstone with white plaster panels. This
is lighted by means of line source tungsten lamps (approximately
25 watts per foot) the reflectors being located above the mould-
ing running around the cove. Fourteen opalescent glass bowls;
equipped with clusters of three lamps each, hung from the ceiling
by long brass rods, furnish a feature which seems desirable, vis.,
a visible source of illumination. The book racks and balconies
in the side bays are lighted by 60-watt clear tungsten lamps in
opalescent bowl-shaped reflectors. Paintings on the rear wall
are lighted by individual mirrored trough reflectors equipped with
25-watt clear tungsten lamps on one foot centers. The entire
Fig. 9. — Distinctive illumination of a shoe store.
Fig. 10.— Distinctive illumination of an antique store.
Fig. ii. — Distinctive illumination of a book store.
pig. 12.— Distinctive illumination of a ladies- wear store.
LAW AND POWELL: STORE LIGHTING
525
front of the store is of plate glass, thus furnishing an excellent
supply of daylight, and the cases and shelves being of light oak
give the room a most cheerful appearance.
Windows
Store (main bay)
Length 98 ft. (29.87 m.).
Width 28 ft. (8.53 m.).
Area 2,740 sq. ft. (254.5 sq. m.).
Ceiling height (maximum) 30
ft. (9.14 m.).
Lamps 9 ft. (2.74 m.) from
floor.
128 35- watt tubular tungsten
lamps.
42 40-watt clear tungsten lamps.
Total watts (approx.) 7,580.
Watts per square foot 2.8.
No special lighting, as the win-
dows extend to the top of the
arch and the whole store is a
flood of light.
LADIES' WEAR.
This type of shop is really divisible into two classes: namely,
general and specialized.
As an example of the first class, the Fifth Avenue (New York)
store of J. M. Gidding, which is shown in Fig. 12, may be given a
little attention. The lighting units are of the sunburst" type,
consisting of 6 regular and 6 round bulb, all-frosted tungsten
lamps below a gilded composition plate, all suspended by a single
chain. The fixtures are pleasingly harmonious with the cream
colored ceiling and delicate gold lining. Show cases and wood-
work are of magnificent Circassian walnut, which blends well
with the rich carpet of green and tan. The wall visible above
the dressing room is largely covered with gilt figures.
Window
Mirrored trough reflector with
25-watt tungsten lamps on 9 in.
Store
Length 55 ft. (16.76 m.).
Width 48 ft. (14.63 m.).
Area 2,640 sq. ft. (245.25 sq. m.).
Ceiling height 12 ft. (3.66 m.).
Lamps 9 ft. (9.74 m.) from
floor.
90 20-watt all-frosted tungsten
lamps.
90 25-watt all-frosted tungsten
lamps.
Total watts 4,050.
Watts per square foot 1.5.
centers. White false ceiling,
with two alabaster carved bowls,
equipped with 6 40-watt tungsten
lamps each, are suspended from
this. Base and trim of window
Circassian walnut.
526 TRANSACTIONS I. £. S. PART II
The specialized ladies' wear shop is exemplified by the shop
of W. B. Crocker, Fig. 13, which handles mourning goods ex-
clusively. The scheme of decoration is very appropriate. As one
observer remarked, "A sombre cheerfulness fills the room." A
rich gray carpet is but a shade darker than the wall covering,
which, in turn, matches the woodwork of the show cases and trim.
The chairs are of gray oak and tables of wicker work.
The ceiling is white, and suspended from this by long chains are
four seven-light and one fourteen-light shower fixtures, dull sil-
ver finished. Low wattage, bowl-frosted tungsten lamps are used
with diffusing shades.
Under the balcony at the rear of the store are full-length mir-
rors. Localized illumination is provided at each by a 25-watt
all-frosted round bulb tungsten lamp. In the front portion of the
store, the general illumination is supplemented by two-arm
brackets similar in finish and equipment to the overhead units.
Store Window
Length 76 ft. (23.16 m.) . Mirrored trough reflectors
Width 16 ft. (4.88 m.). equipped with 50-watt Gem
Area 1,210 sq. ft. (112.4 sq. m.). lamps on 9 in. centers. Wood-
Ceiling height 18 ft. (5.48 m.). work gray. Mirrors at side.
Lamps 10 ft. (3.05 m.) from Beaded crystal hemisphere set
floor, in center of the ceiling.
60 15-watt bowl-frosted tungsten
lamps.
18 25-watt round bulb tungsten
lamps.
Total watts 1,390.
Watts per square foot 1.15.
STATIONERY.
When mention is made of this class of store, one involuntarily
pictures in his mind a low-ceilinged, dingy room with everything
arranged in a haphazard manner; cigars, newspapers, candy and
stationery in a grand mix-up. In contrast to this, is is pleasing to
note Dennison's new store on Fifth Avenue (New York), Fig. 14.
Immense square columns support a pure white ceiling beautifully
decorated with raised plaster figures. The woodwork of shelves,
drawers, show-cases and counters is of carefully selected
weathered oak. Neatness is the predominating feature, and the
semi-indirect lighting units of canary and white glass harmonize
5^
Fig. 13. — Distinctive illumination of a ladies' wear store.
Fig. 14.— Distinctive illumination of a stationery store.
Fig. 15. — Distinctive illumination of a restaurant.
Fig. 16. — Distinctive illumination of delicacy store.
LAW AND POWELL: STORE LIGHTING 527
perfectly with this characteristic. The single-chain suspension
and the bowl itself are designed along lines of simplicity. Six
lamps are installed in each fixture.
A balcony is seen at the rear of the store ; this is used for
office purposes and is lighted by four four-light units with short
ceiling suspension similar in design to the large units.
The space below the balcony has been given quite a bower-like
appearance by the use of a false ceiling of green lattice work and
a profusion of paper flowers.
Store Window
Length 67 ft. (20.42 m.). Mirrored trough reflector equipped
Width 36 ft. (10.97 m.). with tungsten lamps 25 watts
Total area 2,410 sq. ft. (223.88 foQt
sq. m.).
Ceiling height 18 ft. (5.48 m.).
Lamps 12 ft. (3.66 m.) from
floor.
60 100-watt clear tungsten lamps.
Total watts 6,coo.
Watts per square foot 2.5.
RESTAURANT.
There seems to be no definite practise with regard to the light-
ing or rooms of this nature. Some proprietors desire a great
flood of light and the attendant sparkle as produced by crystal
chandeliers; others demand a soft, well-diffused, low general
illumination supplemented by localized table lamps. Bergfield's
restaurant, on Broadway, New York, Fig. 15, is an example of
lighting with the latter idea of proper lighting. Totally indirect
single unit, mirrored reflector lighting units of composition
moulded into an Egyptian design are used. Mirrors are set into
the wall panels, and above each is a two-arm verde finished
bracket with low-wattage multiple lamps and silk shades. The
wood work is cream, with gold decorations; ceiling white; wall
panels old rose, satin finish; chairs mahogany finish, and the
carpet a neutral green.
Store Store
Length 77 ft. (23.47 m-)- 5 250-watt clear tungsten lamps.
Width 27 ft. (8.23 m.). 4 100-watt clear tungsten lamps.
Area 2,080 sq. ft. (193.23 sq. m.). 68 10-watt clear tungsten lamps.
Ceiling height 13 ft. (3.97 m.). Total watts 2,330.
Lamps 9 ft. (2.74 m.) from Watts per square foot I.I.
floor.
528 TRANSACTIONS I. E. S. — PART II
DELICACIES.
The Specialty Shop in Boston, Fig. 16, has a dark wainscoat-
ing about 6 feet (1.83 m.) high; above this the walls are divided
into panels ; in each of these panels is set a heraldic design, which
has been adopted by the proprietor as a trade mark. The ceiling
is white, glazed, and divided into polygons by the moulding. The
show cases and counters are divided into panels which are prac-
tically replicas of the wall panels. The floor is mosaic tile. The
lighting system is remarkably in accord with the general scheme.
Between adjacent panels is a torch fixture with an upright lamp
and diffusing ball. Counter standards, wall brackets and four-
arm ceiling fixtures carry bowl-frosted tungsten lamps and pyra-
midal art glass shades which are finished to match the counter
and wall trim. A most pleasing harmony is secured.
As proof that original store layouts are noticed by the general
public, the management of this store reports that the lighting
is the subject of many favorable remarks, both by local and
out-of-town customers.
Store.
Length 72 ft. (21.94 m-)-
Width 40 ft. (12.19 m.).
Area 2,880 sq. ft. (267.55 sq. m.).
Ceiling height 20 ft. (6.09 m.).
Lamps t8 ft. (5.48 m.) from
floor.
25 60-watt tungsten lamps.
14 25-watt tungsten lamps.
Total watts 1,850.
Watts per square foot 0.65.
CONCLUSION.
A sufficient number of individual installations have been de-
scribed to indicate quite clearly that the lighting system should
blend with the general scheme of decoration. The illuminating
equipment, rather than being dazzling, glaring or commonplace,
is inconspicuous, and forms a part of the furnishing of the room.
It must be borne in mind that the methods outlined above are
not the only correct schemes of lighting to use; often in the lay-
ing out of an installation the ideas or desires of the proprietor
LAW AND POWELL: STORE LIGHTING 529
will produce considerable deviation from the scheme which would
be most in keeping with the period of architecture that is being
followed.
From the descriptions given, it can be seen that one is able to
apply the commercial diffusers and reflecting devices to almost
any class of service. As far as possible the endeavor has been
made to discuss stores which had standard equipment, thus show-
ing that there is no necessity for the design of special auxiliaries.
Expanding this idea slightly, the authors believe that the stores
described are distinctive and yet, with the exception of the carved
alabaster bowls, the initial cost is relatively low.
These illustrations might have been continued at great length
and an appropriate use found for almost all the equipment listed,
but this is obviously out of the question, so the paper can well be
closed with the admonition. — In designing the lighting for shops
of the class treated in this paper, use discretion in the selection
of lighting units and do not offer the prospective customer some-
thing which is, on the face of it, purely utilitarian.
The authors desire to thank the photographic bureau of the
New York Edison Company, for their assistance in taking pho-
tographs and making autochromes and lantern slides.
DISCUSSION.
Mr. M. H. Flexner: Knowing that the larger units are
more efficient than the smaller ones; that just as good results
can be accomplished with the larger units — and I am satisfied
that equal artistic effects can be obtained as with the smaller
ones — I would like to ask Mr. Powell why the clusters seem
to be so much in evidence?
Mr. S. G. Hibben : It seems from the foregoing paper that
most of these distinctively lighted stores have had their lighting
fixtures built up to be in harmony with the interior decoration.
It would be excellent to have fixtures and surrounding decora-
tions planned and built up simultaneously, as is now being done
by some of the large department stores. This brings forth the
advantages of co-operation between the lighting engineer, and
the architect, or particularly the interior decorator.
530 TRANSACTIONS I. E. S. PART II
I believe that of the five or six problems of residence light-
ing that I come in touch with every day, there are only perhaps
one or two that are new installations, and it seems to me that
the illuminating engineer is called on for advice only when the
room or building is so poorly lighted that something must be
done. He is a sort of "lighting doctor," giving a cure rather
than a preventative.
It may be of interest to notice how many types of glass
appliances are regularly available for distinctive store lighting.
Quite often the consumer may make large expenditures for a
peculiar or special design of glassware, that might be saved
him, were he more thoroughly acquainted with the large variety
of illuminating glassware on the market. In briefly mention-
ing some of these available types, I would call attention to the
rapidly increasing number and variety of glass bowls and semi-
indirect reflectors, with open tops, or covered with crystal glass
plates, or partially closed as in urn shapes.
The decorator can choose from a large variety of period de-
signs, Gothic, Elizabethan, Doric, Adam, Georgian, William
Morris, Colonial, etc. One can have glass cylinders, columns
with bases and capitols, or glass troughs for outline and cornice
lighting. Flat or configurated diffusing glass plates are avail-
able for ceiling panels, and plaques for side walls that may serve
to replace open bracket lights.
The painting or color decorating of glass is another feature
being developed to bring out a relief design by shading, or to
give special monograms in glass for fraternal orders, clubs or
stores that feature a trade design or coat-of-arms. In distinctive
stores, like these described, the gold or silver fixtures, or ones ,
with such finishes may be matched by properly colored glass-
ware.
Mr. R. B. Ely: A paper on distinctive store lighting I think
should be encouraged. In some of the illustrations, particularly
the restaurant where indirect lighting was employed. I notice
that some direct lighting brackets were used. These brackets,
I think, would be a distracting feature. In another of the
installations, I notice that one lamp has been exposed; this is
a particularly bad feature in the system. Take the installation
STORE LIGHTING 531
in the tea room with the silk shades and the wiring, which I
believe was exposed. This artistic installation must have been
comparatively inexpensive. I also notice a lack of portable
lamps in the installations shown. Such lamps I think are com-
ing into more general use. The greatest drawback to their use
is the care of the extension cords. However, in some instances
where lamps of this kind have been equipped with leaded glass,
and other artistic designs, they have been very effective.
Mr. A. L. Powell: Referring to Mr. Flexner's query as to
the cause of the clusters being used to such a great extent in
the examples shown, I would say that the majority of the
installations were designed by the architect. My experience
has been that the architect prefers the clusters to the individual
lamp for the additional flexibility possible with it, and also for
the reason that if one particular lamp burns out, the illumina-
tion in a given section will not be materially decreased.
The advisability of combining direct and totally indirect light-
ing in the restaurant was questioned. It seems that the use of
brackets in this particular case, is very feasible, for a relatively
large percentage of the patrons are ladies, and they prefer light
coming from the side, for the illumination of their faces. In
the case in hand, rather dense, shirred silk shades were used,
and the glare was absolutely unnoticeable ; in fact these units
added considerably to the pleasantness of the room.
The use of portable lamps was suggested as an advisable
feature of distinctive store lighting. Careful reference will
show these in use in Page & Shaw's and the Boston Specialty
Shop, and in Schraffts the counter lamps perform the same
function.
Mr. Hibben's remarks, as to the use of tinted opalescent
glasses, is very timely and there are many cases in which they
fit in excellently with the general architectural schemes, and
beautiful effects may be produced with them.
The lack of halation of the autochrome plate may be readily
explained. The ordinary plate is used with the emulsion side
toward the lens; light passes through the emulsion, strikes the
glass plate, and being reflected, re-enters the emulsion, pro-
ducing halation. The autochrome plate is placed in the camera
532 TRANSACTIONS I. £. S. — PART II
with the glass side toward the lens and a piece of dull black
cardboard against the emulsion; light passes through the plate,
and the emulsion and strikes the dull black surface of the card
which reflects but very little light, thus reducing halation.
HARRISON AND EDWARDS \ INCANDESCENT LAMPS 533
RECENT IMPROVEMENTS IN INCANDESCENT LAMP
MANUFACTURE.*
BY WARD HARRISON AND EVAN J. EDWARDS.
Synopsis: i. Increased mechanical strength of tungsten filaments:
The strength of tungsten filament has more than increased 300 per cent,
since 1908, and the strength of drawn wire has increased 40 per cent,
since 191 1. Greater strength permits operation at increased efficiencies
at no decrease in total life. 2. Better candle-power maintenance: The use
of chemical in the bulbs which has become general during the past year
has reduced the blacking of lamps to a marked degree and it is therefore
possible to operate them at efficiencies correct from the standpoint of
total life with no shortening of the useful life. The performance of
chemical lamps is not satisfactory when operated at low efficiencies.
3. Decreased bulb size: Use of chemical has made possible a substantial
reduction in bulb size for several lamps. Decreased bulb size reduces
manufacturing costs and broadens the application of the lamp. 4. Stand-
ardization : During the past year lamp dimensions have been standardized
in every particular. The average deviation from the standard is less than
one-fourth that of a year ago. 5. Helical filaments : The introduction of
the coiled filament makes possible many new forms of lamps which here-
tofore could not be manufactured. The strength of the filaments is
increased by this process and the candle-power maintenance is not affected.
The operation of helical filament lamps at high efficiencies and their use
in small bulbs rather than poorer performance are the causes of their
comparatively low life ratings. The new tubular lamp and the focus type
lamp have many applications such as showcase lighting, use in projec-
tors, stereopticons and the like.
The purpose of this paper is to review, briefly, recent improve-
ments in the art of incandescent lamp manufacture and their
commercial applications. The more important of these improve-
ments may be grouped under five heads :
(1) Increased mechanical strength; (2) Better candle-power
maintenance, obtained by the use of chemical in the bulb; (3)
Decreased bulb size; (4) Standardization of lamp dimensions;
(5) Production of filaments in helical form.
* A. paper read at the seventh annual convention of the Illuminating Engineering
Society. Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
II
534
TRANSACTIONS I. E. S. — PART II
STRENGTH.
The earliest tungsten-filament lamps were so fragile that,
barring those infrequent cases in which the lamps blackened after
the first few hours of service, their performance was judged
almost entirely on the basis of total life figures. It is well known
to the members of this Society that the strength of the pressed-
filament lamps steadily increased and with the introduction of
the drawn-wire lamp in 191 1 a very marked improvement took
place. Perhaps some do not realize, however, that the increased
strength of drawn wire which has been effected during the past
two years is even greater than the difference in strength between
the pressed-filament and the drawn-wire filament of 191 1. That
this is actually the case is shown graphically in Fig. I which gives
Z090
^£60
So
§fc50
5"
«z40
>i30
5^20
^° 10
1907 1908 1909 1910 1911 1912 1913 1914
YEAR
the result of transverse tests of filaments manufactured in each
year since 1908. The ordinates of this curve are proportional to
the distance through which a filament of given dimensions will
bend before breaking when stressed by a gradually increasing
load. An increase in the strength of a lamp also implies a more
homogeneous and uniform filament and the practical result is that
the lamps may be operated at a far higher efficiency than before,
with no decrease in total life.
CANDLE-POWER MAINTENANCE.
The other factor most important in determining the useful life
of an incandescent lamp is the decrease in candle-power with
age which takes place as the result of the blackening of the bulb
by particles thrown off by the filament. While steady improve-
<K
6^
120
110
100
BE
£90
£80
370
zfeO
°50
g«
o: JO
u
°- 20
10
0
->
^
^
....
-2[
-
--
-— .
~"""*&a
L*
*B.a
B.Q.
H0UR6 LIFE
tL-*fcJU*J
Fig. 3-
*.See lines 12 to 21 011 page 547.
« BOOK .CHED. T-l
S3 Mazda
e Straight Side
le Large
-"■60
™ 100-130
= 1.12
» S-21-E2
e #101
UCTION FACTOR ^rf
. OVERALL LENGTH X4O mil 5£"
=ENTER CFROM BABE) g? jm 3-7/I6"
:enter ip,o» ,ul, .no. 1 43nmi.il/16nl
• top tube 5* or 6 mm e
*TH TOP TUBE §9 BOB ft "
'»--»'». 3 ™ (2f-3i) 7
. O.AM. TOP BUTTON 5*. mm UM"' ,
D,«. BOTTOM BUTTON Q j^ 20/64"^?
TOP SUPPORT. 6 ' ^
SUPPORT DIAM. 4^ mils
support hook 4^-32-7-3 (closed)
BOTTOM SUPPORTS ^
OM SUPPORT DIAM I3 mils
OM SUPPORT HOOK 1J_26_8_3 (olCSed)
mum spread 29 mm 72/64 "
""""" 6 mm (4-8)
>ING B0# 1
™ No. 14
er position All bottom hooks
I . BTEMTUBINO H mm (lO^-lli^
""" 19 mm f
STEM LENGTH 2J TJVm 1"
KNESS STEM SEAL Ji mm
' 507-I6-4-I656 Copper Tube
BULB NECK jj « mm I-5/16"
T FORMULA fJO.2
.USTSCHEO. HO.1 (5-5-2)
o Sohed. Ho. 2
PKG.OUAN. 100
PKS.WE.GHT 33 lb8.
PPE" A2145
R CONTAINER ^219 5
:R CONTAINER M2142Q
52-22M-60
Sheet Bo. 153
Fig. 4.
B.
Fig- 5-
HARRISON AND EDWARDS : INCANDESCENT LAMPS 535
ments have been made since 1908 in the direction of diminishing
this black deposit, except in the smaller sizes, they have in no
way kept pace with the increasing strength and uniformity of
the lamp filaments and, owing to this blackening of the bulb,
which becomes greatly accelerated at high filament temperatures,
it has been found impossible to operate many of the lamps at an
efficiency warranted by their total life performance. To be of
practical value, therefore, further improvements in the quality
of incandescent lamps must necessarily take place in the direction
of better candle-power maintenance ; hence this phase of the
problem has been given particular study during the past two
years.
A long series of laboratory experiments led to the introduction
of a chemical in the bulb which, under proper conditions, will
combine with the black deposit in such a manner as to render it
light in color and thus reduce the bulb absorption to a marked
degree. For example, the lamp whose performance is shown by
curve A in Fig. 2 has a life sufficient so that it might well be
operated at an initial efficiency of 1 watt per candle instead of
1.1 watt per candle, its actual rating. However, if burned at the
former efficiency, its candle-power life performance would
approximate curve B, and at the end of the first few hundred
hours of burning, the lamp bulb would be darkened to such an
extent that a discriminating user would become dissatisfied and
would complain of the short useful life of the lamp. Others,
not so careful, would perhaps keep the lamp in service, but at
the expense of a serious loss in economy due to the constantly
increasing cost of energy per candle-power hour. By the use of
the chemical referred to above, the candle-power life perform-
ance of the lamp can be improved in accordance with curve C
and its useful and total life are made nearly identical. However,
if the lamp supplied with chemical were burned under voltage
so as to operate at an efficiency as low as 1.1 watts per candle, its
performance would not be that shown by the dotted curve D, for
the .chemical will not operate properly except at high tempera-
tures; at an efficiency of 1.1 watts per candle, the performance
of the lamp with chemical would be but little, if at all, superior
to that of a lamp not so equipped.
At the present time all lamps above the 40-watt size are sup-
536 TRANSACTIONS I. £. S. PART II
plied with chemical. There is no particular advantage in intro-
ducing the chemical into smaller lamps inasmuch as the useful
life is still limited by failure of the filament rather than by bulb
blackening.
DECREASED BULB SIZE.
Aside from bettering the life performance of the tungsten-
filament lamp, much time and effort have been directed toward
decreasing the cost of the product and broadening its application
The cost of manufacturing and handling the lamps varies almost
directly with the bulb size, and the cost of reflectors and similar
accessories increases in an even greater ratio ; hence, from the
standpoint of economy, a substantial decrease in bulb size is
equivalent to a considerable increase in life performance or effi-
ciency. Fig. 3 shows the relative sizes of the old 60-watt lamp
with the skirted base and the lamp as at present marketed in
the S-21 bulb. Owing to the use of chemical in the new 60-watt
lamp, it will give as long a useful life as the older type at a
decrease of 25 per cent, in renewal cost. The new lamp can also
be used in many locations where the size and appearance of the
older type rendered it inapplicable. The demand for a 60-watt
tungsten-filament lamp in a smaller bulb was probably more pro-
nounced than in the case of any other size, inasmuch as this is
the highest wattage which would be used to replace the ordinary
carbon lamp unit for unit. At the same time, however, there is
a well-marked tendency toward a decrease in size for all lamp
bulbs.
STANDARDIZATION.
Of particular interest to the illuminating engineer are the efforts
which have been put forth recently toward the standardization
of lamp dimensions. Fig. 4 is a factory specification sheet for
the new 60-watt lamp, for which forty-seven distinct items have
been standardized. The most important standard dimension from
the viewpoint of the illuminating engineer is the distance between
the center of the light source and the base contact of the lamp,
since it is this dimension which has the greatest effect upon the
light distribution with various reflectors. The average deviation
from the standard in this dimension is now less than one-fourth
of the average deviation found a year ago.
Fiar 6.
[111! Will
«T_
Fig. 7-
Fig. 9.
p . - PPI
^■t^SS::-. J]"?
"HffiSil i&k
HARRISON AND EDWARDS : INCANDESCENT EAMPS 537
HELICAL FILAMENTS.
Perhaps the most far-reaching of the improvements which have
taken place during the past year is the general introduction of
the coil filament type of tungsten lamp. It is well known that
after the voltage, wattage and efficiency of an incandescent lamp
have been determined, all of the filament dimensions are fixed.
For example, the old pressed-filament i io-volt, 40-watt lamp when
designed to operate at 1.25 watts per candle had a filament diam-
eter of 1.605 mn"s and a length of 21.25 inches (53.975 cm.).
The lamp bulb and filament supports were necessarily of such
size and shape as to take care of this length of filament. It was
not possible to obtain a concentrated filament lamp, neither could
one having a single line of light be manufactured unless the latter
were placed in a bulb 22 inches (55.88 cm.) long. Recently it
has been found that drawn wire filaments of all sizes can be
coiled into the shape of a helical spring ; this greatly reduces their
overall length and makes practicable a lamp of almost any form
desired. This process was first developed in connection with the
low-voltage, high-current auto headlight lamps, and later was
found practical for even the smallest filament. The diameter of
the helical coil is ordinarily not more than seven times the diam-
eter of the filament itself and, therefore, the difference in poten-
tial between successive turns is very small, usually about one-
tenth of a volt. At this low voltage there is practically no tend-
ency for the current to short-circuit its regular path, even when
the coils appear to touch each other.
In addition to the above, it has been found that coiled filament
lamps are much stronger than those of the standard type, quite
as strong, in fact, as the old carbon lamps. This is due in part
to the greater ability of the coiled filament to absorb shock and
partly because tungsten wire seems to be increased in strength
by stressing beyond its elastic limit.
Fig. 5 shows three new types of lamps standardized since the
introduction of the coiled filament. Fig. 5-A is a no-volt,
15-watt lamp having a 1 9/16-inch (39.68 cm.) bulb with cande-
labra base, and illustrates the possibilities in small high-efficiency
light sources.
Fig. 5-B illustrates a lamp with a bulb 1 inch (2.54 cm.) in
diameter and 12 inches (30.48 cm.) long that is made in the
538
TRANSACTIONS I. E. S. — PART II
25 and 40-watt sizes and is intended primarily for showcase
lighting, but which will no doubt find a number of other applica-
tions. For example, Fig. 6 shows this lamp in a frosted bulb, so
placed over a dresser that it serves as a satisfactory substitute
for the two wall brackets often located at either side of the
mirror.
Fig. 7 shows a fixture which has been manufactured for use
with this lamp in showcase lighting. Its distinctive features are
that the socket is supported in such a manner that the lamp can
be swung into the position shown, in order to facilitate renewal
175' 165° 155'
25° 35° 45°
REDUCTION FACTOR
100 WATT REGULAR Rt>. BULB
IOOWATT FOCUS FILAMENT NO. 1
100WATT FOCUS FILAMENT NO. 2
Fig. 8.
81
98
89
or the cleaning of the reflectors; that the current is carried
through successive reflectors by dowel pins fitting into receptacles
at the ends of each unit and by wires placed in metal moulding
at the back ; and that means are provided for securing the reflec-
tor to standard showcases of all types. This unit is assembled
complete before leaving the factory and, therefore, the installa-
tion can be made at a minimum labor cost. "Blanks" or "spacers"
HARRISON AND EDWARDS : INCANDESCENT LAMPS 539
are provided where the high intensity which would be obtained
from a continuous line of lamps is not desired. End supports
are also manufactured to carry the wires up from the floor of
the showcase and to secure them to the first reflector.
Fig. 5-C shows a ioo-watt concentrated filament lamp of the
focus type which should fulfill a variety of needs. The coiled fila-
ment lamps even in this form will give fully as good a life per-
formance as lamps of the regular type when operated under simi-
lar conditions. In most cases, however, a high intensity from a
very small source is desirable and, for this reason, they are
operated at what would be considered over-voltage for the stand-
ard lamps and are also frequently placed in smaller bulbs in order
to permit their use with reflectors and lenses of moderate dimen-
sions. The focal length of a lens or reflector cannot be less than
one-half the diameter of the bulb, if parallel rays of light are to
be secured. In consequence of these facts, the life ratings for
focus type lamps are considerably lower than for the correspond-
ing standard no-volt units. In rating coiled filament lamps of
all types, efficiency and candle-power values must of necessity be
based on total light output rather than upon the intensity in any
one direction, for their reduction factor is much higher than for
ordinary lamps and is also far from a constant quantity. Fig. 8
shows the distribution curves of three ioo-watt lamps, which
illustrate the latter point.
The ioo-watt focus type lamp will successfully operate small
stereopticons for lecture room or residence use and is especially
convenient for demonstration work in connection with college
courses. It is perfectly steady and requires no rheostat and no
attention other than switching on and off. With these lamps it is
possible to obtain a brilliant illumination on a small screen for a
comparatively low energy consumption.
Fig. 9 illustrates a problem in sign lighting which was readily
solved by means of concentrated filament lamps. The use of 250-
watt regular lamps in angle reflectors located 4 feet out from the
upper edge of the tank was first considered, but it was found that
this system would not give satisfaction either from the stand-
point of uniform illumination by night or of appearance by day.
Two ioo-watt, no-volt lamps in 12-inch (30.48 cm.) silvered
parobolic reflectors placed in a weather-proof housing at a dis-
540 TRANSACTIONS I. E. S. — PART II
tance of approximately 50 feet (15.24 m.) supplied the illumina-
tion for the portion of the tower illustrated in the photograph.
Two more units are used to illuminate a similar sign on the oppo-
site side of the tank.
Fig. 10 shows the fagade of the Engineering Department Build-
ing, Cleveland, Ohio, as it will appear from Euclid Avenue when
illuminated by two banks of projectors, each of which will con-
tain nine 250-watt concentrated filament lamps of the focus type.
Reflected light from this building is utilized for the illumination
of the grounds immediately adjoining.
DISCUSSION.
Mr. Norman Macbeth : There is one point I would like to
bring in here in connection with the discussion on standardiza-
tion which arose earlier in this meeting. It is agreed that the
information bearing on the relative position of the filament in
an incandescent lamp and of the lamp in the reflector is some-
thing that engineers require in all reports of photometric inves-
tigations. I had occasion some time ago to go over the standard
specifications of four laboratories, and in checking them up I
found that their "a, b, c, d and e" which referred to these
relations showed a serious lack of agreement in some, one or
all of the various mentioned designations.
It ought to be the duty of the committee of this society to
see that these a, b, c, d and e designations have the same mean-
ing with all of these laboratories, and it would also be desirable
to provide for other glassware and fixture conditions than those
now covered by these letters which refer only to simple reflec-
tor designations.
I would like to raise the question on this 80 per cent, so-
called smashing point which I remember Dr. Carl Hering
stated at one of the earlier meetings of the Philadelphia
Section he had determined many years ago by himself,
and that same was now a matter of record in the proceedings
of the American Institute of Electrical Engineers This was
the point at which when considering an average cost for the
lamps and for the energy used, a lamp having deteriorated to.
80 per cent, of the initial candle-power was no longer econom-
ical ; that beyond this point the total costs in relation to the
INCANDESCENT LAMPS 541
light produced resulted in a new lamp being less expensive.
From statements that have been made applying this 80 per cent.
to the tungsten lamp it would appear that we have lost sight of
the calculation of which this 80 per cent, was the result. It
can hardly be possible that with the more expensive tungsten
lamps and the very much lower energy cost per light unit that
this figure would be true for these lamps.
I would like to ask the authors the probable hour's life rep-
resented by the curves in Fig. 2. While this information is not
really necessary to bring out the point for which these curves
were used information as to the total hours represented by the
two curves would considerably widen the use of this one illus-
tration.
Mr. W. F. Little : The standardization of filament dimen-
sions and of the location of light center in incandescent lamps
is perhaps of more importance to the lamp purchaser than to
the manufacturer, as the variation from standard conditions
might be such as to change the characteristic of an intensive
reflector to that of an extensive reflector. As representing the
purchaser, the Electrical Testing Laboratories has for some
time past urged close adherence to the dimensions as laid down
by the manufacturer, and has included dimensional requirements
in its inspection criteria.
Mr. G. H. Stickney : There should be no loss in efficiency
from light falling upon filament surfaces. Any heat trans-
mitted from one part of the filament to another in this manner
would simply go into raising the temperature of the filament
and therefore be returned as light. (On account of the thin-
ness of filaments and the heat conduction of the material, it
would hardly be possible for one side of the wire to attain a
perceptibly higher temperature than the other.)
We cannot over-estimate the importance to the art which
may be derived from our ability to concentrate the filament, as
referred to in the paper. As referred to in previous discussion,
this has introduced a considerable number of new applications
for incandescent lamps and created some entirely new fields of
lighting. The importance of this has been appreciated for some
time and we have made a very considerable study of the uses
542 TRANSACTIONS I. E. S. — PART II
of point sources of light. In the long run, however, I am not
sure that the point source may not prove of even greater im-
portance in handling problems of interior lighting, since it offers
an opportunity for improved control of the light in obtaining
new and desirable effects.
Mr. L,. C. Porter: In speaking of the applications of the
new focus type lamp, there are three fields which seem to open
up considerable use for this lamp. One is in theater lighting.
Some of the larger sized lamps have been used for flood lights
in theaters. By the use of such lamps and large parabolic
reflectors we can get effects which are hard to obtain with the
arc. The incandescent lamp can be very easily and evenly
dimmed or brought up to full candle-power. Focus type lamps
have been used with parabolic reflectors to throw a sheet of light
over each individual drop curtain, in that way obtaining much
more satisfactory lighting effects than previously were obtained
by the use of the arc. The light is steady and considerable
power economy is obtained by its use.
This lamp is also being used in moving pictures for home
use, to do away with the fire risk and auxiliary apparatus neces-
sary with the arc.
Another field in which a little experimenting has been done
is in signal work. Experiments have been conducted where
signals have been transmitted over a distance of 20 miles
with a concentrated filament tungsten lamp and parabolic reflec-
tor. These signals were read without the use of glasses, both
clear and colored screens being used over the reflectors.
Still another field which is opening up is in headlights, espe-
cially for locomotive headlights. There is at the present time
considerable agitation throughout the country on the question
of locomotive headlights. Several states have passed legisla-
tion requiring locomotives to carry more powerful headlights ;
in others this subject is pending. Some of the roads, especially
those with double tracks equipped with block systems, object
to using the very powerful arc headlights. Between the arc
and the oil lamp comes the incandescent. It seems very highly
probable that a 6-volt focus type of incandescent lamp (oper-
ated either by storage batteries or by a small 6-volt turbo-
INCANDESCENT LAMPS 543
generator outfit) will have large application for locomotive
headlights in the near future.
Dr. M. G. Lloyd: The efficiency of an incandescent fila-
ment depends upon the temperature of the surface which is
emitting radiation. The question arises as to whether with a
helical filament the surface temperature can be as high as with
a straight filament having the same life; or whether the life
can be as long for the same surface temperature. In a straight
filament the highest temperature is in the interior of the wire
and it is only the temperature of the surface which affects either
radiation or evaporation. In a helical filament the highest tem-
perature will probably be found on the surface of the wire in
the interior of the coil. This surface is exposed sufficiently to
permit evaporation, but not sufficiently to give out radiation.
The temperature of the exterior surface will as before deter-
mine the efficiency; whereas, the temperature of the interior
surface will probably determine the life. For the same tem-
perature of the external surface it would, consequently, seem
possible that the evaporation would be greater, and hence the
life of the filament shortened as compared with a straight fila-
ment.* Practical experience seems to indicate that this effect
is not appreciable.
Mr. H. Calvert : Referring to Fig. 2, I note that the figures
representing the hours life have been omitted. I think it would
be much more interesting and instructive if these figures could
be inserted. The question of the blackening of the lamp bulbs
has recently taken on a new and interesting phase. A number
of the central stations have recently adopted the policy of giving
free renewals on certain sizes of tungsten lamps. Formerly,
when the consumer had to purchase each lamp he would keep
it in use until it got so black that he could get but little light.
Now, under the new ruling, he is entitled to free renewals, and
the question arises, at what point in the life of the lamp is the
company justified in giving a new lamp? It is generally not so
much the decrease in the actual candle-power of the lamp which
influences him in bringing the old lamp to the central station, as
* Dr. Irving t,angmuir has since stated that in the half-watt lamp using the helical
form of filament the interior surface is at a temperature about 12 degrees higher than the
exterior surface.
544 TRANSACTIONS I. $. S. PART II
it is the blackened appearance of the bulb. I would therefore
like to ask the authors what, in their opinion, is the limit of this
term "useful life" which they use, and at what point, in their
opinion, are the various companies justified in giving free re-
newals ?
Dr. R. E. Myers : This is a very interesting paper which the
authors have given us and there is very little to add. The prin-
cipal features upon which the recent improvements depend have
been given in a very able way.
Greater emphasis might be laid on the fact that the latest
developments in the standard lines of tungsten lamps, which prob-
ably represent over 90 per cent, of this product, are due to
improvements in the filament and to the use of chemicals.
The improvements in filament are of two kinds. First, we are
now using a drawn wire which has a much greater tensile
strength than the pressed filament of the earlier tungsten lamp.
This gives low manufacturing shrinkage and reduced cost.
Second, the present filament is much stronger throughout its
burning than the older filament. This is due to certain changes
of which I am not at liberty to speak owing to trade reasons.
These changes, however, are of the utmost importance in the life
of the lamps of lower wattage of the standard line.
Large improvements in efficiency have certainly been made
by the use of chemicals in the lamp. However, I believe that
still greater ones are yet to come. This field of research will
doubtless prove a very fertile one and I think that in a great
many cases this type of lamp in actual use will prove equal to or
even superior to the new gas filled lamps.
The authors may have to change their opinion regarding the
use of chemicals in the smaller lamps. Recent experiments tend
to prove that certain chemicals can be used to advantage in these
also.
Dr. H. E. Ives : Another application of these helical fila-
ments— not a commercial application — is in photometric research.
In this a valuable consideration is the ability to obtain a wide
range of illumination without resorting to sector disks or absorb-
ing screens. With the ordinary carbon lamp it is impossible to
get nearer the photometric screen than about 10 inches, because
INCANDESCENT LAMPS 545
the inverse square law can no longer be applied. If, however
the light is compressed into these little coils it is possible to
push the light up until the bulb actually strikes the receiving
surface without involving errors from inverse square law cal-
culations which means that one can increase the range of il-
lumination at least ioo times over what was available before.
In this connection I have had, through the kindness of Mr.
Stickney, several of these lamps to try out and I have found
them extremely convenient in so far as the ability to obtain great
variation of illumination is concerned.
One question, however, I have not been able to investigate. I
would like to ask the authors of this paper as to the performance
of these lamps. I suspect that the high voltage lamp would be
apt to be unsteady and unreliable because of the number of fila-
ment supports. The low voltage, however, might be very reliable
indeed. I should like to ask the authors if they have any data
that might enable us to make comparisons as to steadiness with
the usual photometric standards.
Mr. J. R. Cravath : I would like to ask what the effect of
coiling the filament in helical form has upon the efficiency. Is
there any loss of efficiency due to the shading effect of one con-
volution on the next?
Mr. V. R. Lansingh : The use of the helical coil filament
giving a very small body of light will be particularly useful in
the design of reflectors using specular reflection, such as pris-
matic, mirrored, etc. One of the greatest difficulties in a design
of that character is the size of the source of light. The ordi-
nary tungsten lamp filament may be considered as a trans-
parent cylinder, and this introduces a number of problems, one
of which is the shape of prisms, etc., in the case of prismatic
glassware. The ordinary prism, as you probably all know, is
90 degrees. It is possible, however, to design prisms of differ-
ent shapes, so that a higher efficiency is obtained with the pres-
ent type lamp filament than given by the 90 degree prism. For
example, one manufacturer uses a parabolic prism on the out-
side of the reflector.
With the introduction, however, of the new filament, it will
be possible to obtain far higher efficiency than before, and this
54^ TRANSACTIONS I. £. S. — PART II
will be particularly valuable where broad distributions of light
are wanted, such as in street lighting. We may therefore look
probably for a large advance in the design of reflectors from
now on with the new filament lamp.
Mr. L. J. Lkwinson : Referring once again to the light
colored discoloration due to the chemical, I would like to ask
the authors whether the manufacturers consider it necessary
to ship from the factory new lamps, the bulbs of which already
manifest this discoloration to some extent.
Mr. H. S. Dunning : Just a point as to the return of lamps
to central stations : It is probable that for some time lamps of
the 150-watt or larger sizes may be returned because they show
a white deposit on the bulb. It has been our experience backed
by accurate measurements that in most cases the appearance of
this white deposit does not necessarily mean a large decrease
in candle-power, and quite often it will be found that the
candle-power has not decreased at all. Some of these lamps
have a most interesting performance on life test. We have
found cases in which lamps after operating for an extended
period at approximately 100 per cent, of their initial candle-
power seem to lose candle-power for a time and later come back
to approximately normal rating. These changes have been
investigated very closely and it has been found that they are
not due to either photometric or other test errors. I think,
therefore, that a word of caution is in order against the dis-
carding of such lamps, simply because they show an unusual
discoloration of the bulb.
Dr. M. G. Lloyd: The statement of Mr. Harrison that the
smashing point does not depend upon the relative cost of energy
and renewals involves an assumption that should be made clear.
It does not apply to lamps operated at rated voltage, since it is
clear that if the cost of renewal was merely nominal, the smash-
ing point would be reached as soon as the candle-power has
fallen off appreciably, say to 95 per cent. Mr. Harrison's
assumption is that the lamp shall be operated at the most eco-
nomical efficiency and the lower the cost of renewals the lower
should be the specific consumption and the higher the voltage
at which a given lamp should be operated. Since this is a con-
INCANDESCENT LAMPS 547
dition that cannot be conveniently carried out in practise, the
statement that the smashing point is independent of the cost of
renewals and the cost of energy is also one that does not apply
to practical conditions.
Mr. Ward Harrison (In reply) : Mr. Lewison inquires in
what way the characteristics of the coiled filament lamps differ
from those having the standard form of filament. There is a
slight variation in the performance curves, but the principal dif-
ference is in the manufacturing data ; i. e., the length of filament
required, and the diameter for a given candle-power, voltage and
efficiency.
The abscissae of the curves in Fig. 2 were not indi-
cated as actual hours of life because the relative performance
with a given change in voltage is independent of the actual
hours represented by the curve. A typical lamp giving 2,000
hours life to burn out without "getter" would have a life of
about 1,000 hours if the voltage were raised 10 per cent. Under
those conditions the introduction of the "getter" would result
in better candle-power maintenance but the life to burn out
would still be about 1,000 hours. Actual quantitative data on
actual lamp performance were given by Mr. Lewinson.
It has long been considered that the smashing point of incan-
descent lamps for most economical operation depends upon the
rate for energy and the cost of lamps, as well on the form
of the candle-power depreciation curve. More recent investi-
gations show that this smashing point, a certain percentage of
initial candle-power, is determined solely by the performance of
the lamp. The most economical life of the lamp in hours, on
the other hand, is determined by the cost of lamp renewals and
the energy rate, and this life can be secured in practise by
selecting a lamp of the correct efficiency. If the proper choice
of initial efficiency is made, the lamp will be found to have
dropped in candle-power to the smashing point at the end of the
period representing its economical life as determined above. A
full discussion of this subject is given in a technical bulletin
issued by the engineering department, National Lamp Works
of the General Electric Company.
The answer to the question raised by Dr. Ives in regard to
54§ TRANSACTIONS I. E. S. — PART II
using concentrated filament lamps as standards is suggested by
Fig. 8 of the paper. Curves i and 2 are both on concentrated
filament lamps built to fulfill the same specifications and yet it
is seen that the candle-power curves are considerably different.
Inasmuch as a very slight change in the relative position of
filaments will cause a marked change in the shading effect such
lamps can scarcely be used as standards. The filament may
sag slightly after the lamp has been burning for a short period
and thus materially change the distribution curve, although the
total light flux and efficiency would not be affected to any extent.
This well illustrates the necessity of arriving at something
better than simply a watt per horizontal candle-power rating for
incandescent units.
Mr. Macbeth's remarks were most timely as to the desirability
of having some uniform system of designating the position of
the lamp filaments relative to the reflector in recording data on
photometric tests. I am glad to say in this connection that we
have taken the matter up with several other laboratories and
have agreed upon a system of lettering which will be uniform
and which no doubt will eliminate much of the confusion which
has existed heretofore. These designations are of course purely
arbitrary.
Mr. Cravath and Dr. Lloyd spoke about the rise in tempera-
ture of the inner surface of the filament because of the tend-
ency of one portion of the coil to shade another. This effect
so far as has been determined is very slight. As Mr. Stickney
has said, if the radiant energy does not escape at once, it is
merely reflected from the hot surface of the filament one or
more times before passing out of the helix. If this radiant,
energy would be absorbed instead of being reflected, it would
of course raise the filament temperature somewhat, but the net
change would certainly be very slight, since the difference in
temperature between two points on the same cross section of
filament is necessarily small.
There is one more point, and that is in regard to the dis-
coloration of the bulb of "getter" lamps without a material
decrease in light output, which was commented upon by one of
the speakers. This phenomenon is entirely reasonable as in
INCANDESCENT LAMPS 549
those cases the deposit is light in color, similar in fact to an
opal dip or frosting. It may of course have a considerable
effect on the intensity of the lamp in any one direction, but the
decrease in total light flux should not exceed a few per cent.
The tendency of these lamps to be slightly erratic is, of course,
not surprising; however, we believe that the improvement in
this respect during the past few months has been marked.
Mr. J. W. Howell: Most of the things that I had in mind
to speak about have already been mentioned either by Mr. Harri-
son and Mr. Edwards or in the report of your Committee on
Progress, so all I can do will be to give you a little further in-
formation on some of the things which you have already been
considering. In the time in which we live the most rapid advance
in the art of light producing ever known is being made. Even
since incandescent lamps have been made one of their character-
istics has been that the lamps with thin filaments end their life
by breaking, without much discoloration of the bulb, while
lamps with thick filaments and high candle-power blacken the
bulb and become useless before the bulb breaks. So we have
never been able to obtain the good results which we should get
from thick filament lamps. Much progress has been made in
the last two years in preventing the blackening of the bulbs and
thus utilizing the longer life of the thick filament. A year ago
we talked and discussed the action of what we call vacuum
getters on lamps. That same line of work has been conducted
during the past year and very great advances have been made.
There are two reasons why a thick filament lamp gets blacker
than a thin filament lamp : one of the reasons is that it lives
longer and consequently has more time to get black in ; the sec-
ond is the blackening is proportional to the surface of the fila-
ment, and inversely to the size of the bulb. To be properly de-
signed the surface of the lamp bulb should be proportional to
the surface of the filament. But that cannot be. A io-watt lamp
has a diameter of 2% inches, that would require a 100-watt lamp
to be 10 times that diameter of 21 inches; which would be im-
possible.
During the year we have made very great advances in these
large lamps.
550 TRANSACTIONS I. £. S. — PART II
I have just shown a report of some life tests on some hundred-
watt hundred volt lamps. These lamps are tested at 9/ioths of
a watt per candle. Their normal burning efficiency is one watt
per candle. Of these tests only two were completed at the time
the table was made up, which is quite recently. In the two
of the tests the lamps burned a thousand hours at 9/ioths of a
watt per candle and at the end of that time were up to their
initial performance. At the end of 3,100 hours they should show
99 per cent, of their full candle-power.
I also have a report of a test of 250-watt 100-volt lamps. The
lamps were not experimental lamps ; they are not made in the
laboratory ; every lamp was taken from the regular factory stock,
taken without any selection whatever. The lamps were burned
at 24 of a watt per candle on life tests. After 500 hours burn-
ing they were pretty close to 90 per cent, candle-power of initial
candle-power. If they had been burned at their normal effi-
ciency, which is one watt per candle, their life would have been
9.6 as long. So that when they were down to 90 per cent, of
their candle-power, burning on a normal voltage, they would
have burned over 4,000 hours. That shows very great improve-
ment in lamps of that type.
I also have a diagram showing a test on a 400-watt lamp taken
right out of stock, without any selection whatever. The lamp
was started on test in September of last year, just a year ago.
When the diagram was made out, the lamp had burned 6,000
hours and showed about 92.5 per cent, of its initial candle-power.
The lamp when I left to attend this convention had burned 7,000
hours, practically unchanged at this date.
These reports indicate the great improvement which has been
made in the high candle-power lamps. It is the established pro-
cedure in business that as improvements are made the efficiency
is increased. Otherwise the economy of the lamp would not be
realized. It is considered at the present that a thousand hours
of useful commercial life is proper and best, so that as lamps
come to give long lives like that their efficiency is increased so
that the laboratory life is about 1,300 or 1,400 hours, which would
give in commercial practice a life of about 1,000 hours.
There has been under way, during the last two years a very
INCANDESCENT LAMPS 551
remarkable piece of work in the research laboratory of the Gen-
eral Electric Company at Schenectady. The work has been done
by Dr. Langmuir and his assistants, one of whom is here today,
and it was my purpose to tell you a good deal about the work at
this meeting but unfortunately Dr. Langmuir's description has had
to be postponed until next month. A half-watt per candle lamp,
— that lamp is the result of Dr. Langmuir's work. It is a lamp
such as we have here, an incandescent tungsten filament lamp,
the bulb of which is filled with nitrogen. Now the nitrogen has
several effects on the lamp ; some are good and some are bad.
The bad one is that it cools the filament and so reduces the candle-
power. When a gas is introduced into a lamp vacuum it cools
the filament always. In a thin filament lamp the cooling effect
is greater than in thick filament lamps. So the thicker the fila-
ment the more benefit is to be had from the nitrogen gas in the
bulb. In the present state of our knowledge we get our one half-
watt per candle lamp at about 12 amperes; while at 20 amperes
we get 0.4-watt-per candle. Tests of such 20 ampere lamps, have
been made at 0.4-watt per candle. Some of these lamps have
burned 2,000 hours. The large lamp here before you is a 20-am-
pere 112-volt lamp; it consumes about 2,500 watts, and gives
5,000 candle-power. It is supposed to be the largest incandescent
lamp which has ever been made. Unfortunately the lamp has a
little crack in the glass, which has allowed a little air to mix with
the nitrogen ; so that when it is lighted it will show a faint cloud
of tungsten oxide, and it is not an effective lamp. The future
of that lamp, gentlemen, you may all theorize on as much as I
can. You know what the introduction into the art of lamps of
larger sizes than have been made before means ; and the policy
has been established to add to our present line of lamps up to
this 5,000 candle-power size, or higher if necessary. If the
commercial people want a 10,000 candle-power lamp they can
have it ; there is no limit that we know of yet.
That lamp which you see burning is a 500 candle-power lamp,
at 112 volts. Its light is very concentrated and very intense. The
lamp is of a size which does not realize a half-watt per candle:
in fact it is about a 0.6 watt per candle lamp.
Here is a larger lamp one which burns at one-half -watt per
552 TRANSACTIONS I. E. S. — PART II
candle ; it is a good lamp. The air is leaking into that bulb ; its
action will not continue very long, but while it is continuing it
is destroying the filament. I don't know what will happen to
the lamp with the air in it.
There has been another matter which has been alluded to in
the report of the Committee on Progress and also in the paper,
by Messrs. Harrison and Edwards, which I consider very im-
portant and that is what they call a single size wire. It simply
means that the art of drawing tungsten wire has been reduced to
such a fine degree, such an efficiency, that the wire can be drawn
to absolutely the size desired. Of course when you consider the
matter if the dies are proper and of the right size the wire
drawn should be the right size. They are the right size.
And at the present the wire in the filament is of the
right length and diameter; all of the lamps made for a given
voltage are that voltage. If a factory is producing lamps that
are of a given voltage and the photometer disagrees with the
marking, the photometer is wrong. (Laughter.) Gentlemen,
that is a fact. It is true that the grading is better than the photo-
meter. If we find that the lamps test off voltage, in nine cases
out of 10 the difficulty is with the photometer and not with the
lamps. As you know some of our customers in the country buy
their lamps on specifications as to voltage and candle power, and
the lamps submitted during the last six months or more to those
customers are lamps which have never been photometered ; they
are lamps made for a definite voltage and candle power, and the
result is that they are closer to rating than were the previous
photometered lamps.
Mr. L,. J. Lewinson : In order to further emphasize some
of the points brought up in the paper, a table and a diagram are
submitted herewith. The table comprises a summary of the
various watts per candle ratings of tungsten lamps in force in
this country during the past two years. It is noted that with
the exception of the very smallest sizes great strides in efficient
improvement have been made. The diagram shows the very re-
markable improvement life and efficiency of one particular size —
the 250-watt lamp. The heavy horizontal lines represent the
average life value during the periods indicated. It is seen that
INCANDESCENT LAMPS
553
Efficiency Adjustment— Tungsten Lamps.
Watts-per-candle rating-
Wattage April September May
ratings 191 1 1912 1912 1913
15 watts 1. 31 1. 31 1.30 1.30
20 " 1. 31 1.28 1.25 I.25
25 " I-31 1-23 I.I7 I-I7
40 " 1.23 1. 18 1. 17 1. 17
60 " 1. 18 1. 16 1. 16 1. 12
100 " 1. 18 1. 13 1. 13 1.08
150 " 1. 18 1. 12 1. 12 1.03
250 " 1. 13 1. 10 1. 00 1. 00
400 " 1. 13 1. 10 1. 00 1. 00
500 " 1. 13 1. 10 1. 00 1. 00
early in 191 1 the useful life of these lamps was about 700 hours
at 1. 1 3 watt-per-candle. In October of the same year a bulb black-
ening preventive was introduced which had the effect of increas-
— 1
WRC
W.RC. .
1.13
W.P.C.
<U0_
, 1.00
"*
^
\
JAN/12
JULY'15
Diagram A.— Improvement in useful life of 250-watt tungsten lamps.
ing the life to 1,350 hours. By May, 1912, the manufacture of
this type of lamp had been so improved that it was possible to
raise the efficiency, lowering the watts per candle to 1.10, with-
out deleterious effect upon the life. In July, 1912, a new bulb
blackening preventive was adopted, increasing the life to nearly
2,100 hours. In September of the same year a new form of con-
struction was adopted. It was found possible to still further in-
554 TRANSACTIONS I. E. S. — PART II
crease the efficiency, to correspond to i.o-watt per candle with
only a very slight loss in life. In February, 191 3, the form of
construction was again changed, with a resultant life of 2,500
hours. It is seen then that during the brief span of two years
the watts per candle rating has been reduced from 1.13 to 1.10
with an increase of 250 per cent, in useful life. Assuming all
life values to be corrected to one standard watt per candle value,
the inherent improvement in the life of this particular type of
lamps has been nearly 700 hours. All life values quoted above
are to be considered as life to 80 per cent, or prior burn-out.
Illuminating Engineering Society
GENERAL OFFICES : 29 West 39th Street, New York City.
SECTIONS
Chicago Section
CHAIRMAN M. G. Lloyd, 608 S. Dearborn Street, Chicago, 111.
SECRETARY J. B. Jackson, 28 N. Market Street, Chicago, 111.
MANAGERS
Nelson M. Black 1213 Wells Building, Milwaukee, Wis.
J. W. Pfiefer 520 Hannah Avenue, Forest Park, 111.
C. C. Schiller 122 Michigan Boulevard, Chicago, 111.
M. J. Sturm 116 S. Michigan Boulevard, Chicago, 111.
H.B.Wheeler 6204 Lakewood Avenue, Chicago, 111.
New England Section
CHAIRMAN . . C. A. B. Halvorson, Jr., G. E. Co., Center Street, West Lynn, Mass.
SECRETARY . Chas. M. Cole, 156 Pearl Street, Boston, Mass.
MANAGERS
R. B. Hussey .... General Electric Company, West Lynn, Mass.
H. C. Jones 10 High Street, Boston, Mass.
J.M.Riley East Boston Gas Company, Chelsea, Mass.
R. C. Ware 42 West Street, Boston, Mass.
W. E. Wickenden, Massachusetts Inst, of Technology, Boston, Mass.
New York Section
CHAIRMAN .... William Cullen Morris, 124 East 15th Street, New York, N. Y.
SECRETARY . . . Clarence L. Law, 124 West 42nd Street, New York, N. Y.
MANAGERS
H.B.Rogers General Electric Company, Harrison, N. J.
A. S. Ives 84 William Street, New York, N. Y.
H. V. Allen 13 Park Row, New York, N. Y.
W. H. Spencer 239 Tenth Avenue, New York, N. Y.
Oscar H. Fogg 124 East 15th Street, New York, N. Y.
Philadelphia Section
CHAIRMAN . . George A. Hoadley, Swarthmore College, Swarthmore, Pa.
SECRETARY . L. B. Eichengreen, Counties Gas & Elec. Co., Ardmore, Pa.
MANAGERS
H.Calvert 1000 Chestnut Street, Philadelphia, Pa.
F. C. Dickey 30 S. 16th Street, Philadelphia, Pa.
H. H. Ganser . 212 DeKalb Street, Norristown, Pa.
H. A. Hornor . . 102 Hamilton Court, 39th and Chestnut Streets,
Philadelphia, Pa.
Samuel Snyder 1035 Market Street, Philadelphia, Pa.
Pittsburgh Section
CHAIRMAN C. J. Mundo, 1312 Oliver Building, Pittsburgh, Pa.
SECRETARY Alan Bright, 827 Wabash Building, Pittsburgh, Pa.
MANAGERS
H. S. Hower Carnegie Technical Schools, Pittsburgh, Pa.
H. H. Magdsick Nela Park, Cleveland, Ohio.
E. R. Roberts S40 Middle Street, Avalon, Pa.
C. E. Stephens . Westinghouse Elec. & Mfg. Co., E. Pittsburgh, Pa.
S. B. Stewart 435 Sixth Avenue, Pittsburgh. Pa.
Illuminating Engineering Society
LIST OF LOCAL REPRESENTATIVES
State and City Name and Address of Representative
CALIFORNIA: Los Angeles R. H. Manahan,
City Electrician, Los Angeles, Cal.
San Francisco . . . . F. Emerson Hoar,
833 Market Street, San Francisco, Cal.
COLORADO: Denver G.E.Williamson,
Denver Gas & Electric Company, Denver, Colo.
GEORGIA: Atlanta William Rawson Collier,
Georgia Railway & Light Company, Atlanta, Ga.
MINNESOTA: Minneapolis G. D. Shepardson,
University of Minnesota, Minneapolis, Minn.
St. Paul A. L. Abbott,
185 East 4th Street, St. Paul, Minn.
OHIO: Cleveland Ward Harrison,
Nela Park, Cleveland, Ohio.
WASHINGTON: Seattle Fred. A. Osborn,
University of Washington, Seattle, Wash.
TRANSACTIONS
OF THE
Illuminating Engineering Society
Published monthly, except during July, August, and September, by the
ILLUMINATING ENGINEERING SOCIETY
General Offices: 29 West Thirty-Ninth Street. New York
Vol. VIM
DECEMBER. 1913
No. 9
Council Notes.
A regular meeting of the Council was
held December 12, 1913, in the general
offices of the society, 29 West 39th
Street, New York. In attendance were :
C. O. Bond, president; J. W. Cowles,
Ward Harrison, Joseph D. Israel, gen-
eral secretary; V. R. Lansingh, C. A.
Littlefield, L. B. Marks, treasurer;
Preston S. Millar, J. Arnold Norcross,
C. J. Russell, W. J. Serrill and G. H.
Stickney.
The meeting was called to order at
10:35 a.m. by President Bond.
To supplement the sixth paragraph of
the minutes of the October Council
meeting (page 7 of the minutes of the
present administration) the following
amendment was adopted :
The Illuminating Engineering Society hereby
expresses its adherence and support, through
the United States National Committee, to the
International Commission on Illumination; and
as an earnest of its attitude herewith appro-
priates its assigned quota of expenses for the
ensuing year, $100.00, to the use of the said
National Committee.
With the foregoing amendment the
October minutes were adopted.
Mr. Israel reported that the total
membership of the society as of Decem-
ber 10, including applications and resig-
nations to be presented at the meeting,
was 1,401. Counting the applications
for sustaining membership presented at
the meeting there were 31 sustaining
members, which are not included in the
foregoing membership figure. The ex-
penditures for the first two months of
the present fiscal year was said to have
totaled $3,217.31.
Vouchers Nos. 1513 to 1548 inclusive,
aggregating $1,048.46, were authorized
paid subject to subsequent approval by
the Finance Committee.
In accordance with recommendations
contained in a report from the Finance
Committee it was voted (1) to transfer
the account of the society now in the
Lincoln National Bank to the Central
Trust Company at Madison Avenue and
42nd Street, New York, N. Y., provided
interest at 2^ per cent, on the bank
balance cannot be obtained from the
Lincoln National Bank; the Central
Trust Company has agreed to pay the
society interest at that rate on balances
of not less than $1,000 and not more
than $10,000; (2) to increase the
monthly salary of Miss Claire Goldblatt,
an assistant in the office of the society,
from $44 to $54, and (3) to authorize
the printing of another edition of 10,000
copies of the illumination primer,
"Light: Its Use and Misuse."
Verbal reports of progress were re-
ceived from the following committees :
Sustaining Membership, Education, and
Section Development. The reports
were accepted with commentation.
Mr. C. A. Littlefield, chairman of the
1913 Convention Committee, presented
the final report of his committee. The
TRANSACTIONS I. E. S. — PART
receipts and disbursements were said to
have amounted to $1,875 and $1,705.08
respectively. Accompanying the report
was a scrap book outlining the way the
convention was conducted and including
samples of various letters and literature
which had been issued.
Whereupon it was resolved to extend
to the committee a very hearty vote of
thanks for the able and successful man-
ner in which the 1913 convention— prob-
ably the best convention ever held by
the society — was conducted.
Resolved, that a special vote of thanks
be transmitted to the Papers Committee
of the previous administration for hav-
ing arranged for the 1913 convention
an excellent program of papers, which
has been pronounced the best balanced
set of papers ever presented before a
meeting of the society.
A vote of thanks was extended to
Mr. G. H. Stickney for his able presen-
tation of a lecture on industrial lighting
at a session of the International Expo-
sition of Safety and Sanitation held
under the auspices of the American
Museum of Safety in the Grand Cen-
tral Palace.
Mr. G. H. Stickney reported that the
contributions referred to in the fore-
going paragraph had been sent to the
office of the society and that he pro-
posed to submit bills for installing the
exhibit to be paid out of this fund. This
procedure was accepted and accordingly
the Council voted to authorize payment
of such bills up to $300, the amount of
the fund.
It was resolved that a vote of thanks
be sent to the members of the Lighting
Exhibit Committee for their excellent
services in arranging the foregoing ex-
hibit, and to the Electrical Testing
Laboratories for its kind assistance in
the work.
Reports on section activities were re-
ceived from the following vice-presi-
dents: J. W. Cowles, New England;
G. H. Stickney, New York; W. J Ser-
rill, Philadelphia; Ward Harrison
Pittsburgh; and J. R. Cravath, Chicago!
Mr. G. H. Stickney reported infor-
mally on the work of the Papers Com-
mittee of which he is chairman.
The following appointments to com-
mittees were confirmed:
Nomenclature and Standards: C. H
Sharp, Louis Bell, C. O. Bond, E. P.
Hyde, H. E. Ives, L. B. Marks', A S
McAllister, E. B. Rosa; advisory mem-
bers: Andre Blondel, Hans Bunte
Vivian B. Lewes and C. C. Paterson.
Section Development: Alan Bright,
C. M. Cole, L. B. Eichengreen, J. B.'
Jackson, C. L. Law.
Advertising: B. F. Fisher, Jr., F. H.
Gale, J. C. McQuiston and Robert F.
Pierce.
Popular Lectures : Sub-Committee on
Residence Lighting: E. J. Edwards,
chairman; C. R. Clifford, S. G. Hibben,
J. W. Lee and J. L. Wiltse; Sub-Com-
mittee on Store Lighting: A. L. Powell
chairman; J. M. Coles, F. H. Gilpin,'
D. A. Bowen, C. B. Graves and A. S.'
Ives; Sub-Committee on Industrial
Lighting: G C. Keech, chairman; Ward
Harrison, C. E. Stephens, A. J. Sweet
M. H. Flexner, C. A. Luther and H.'
W. Shalling; Sub-Committee on Office
Lighting: C. E. Clewell, chairman;
C. L. Law, J. G. Henninger, T. W.
Scofield.
Papers : Theodore H. Piser and Pro-
fessor W. E. Wickenden.
The resignation of Albert Jackson
Marshall as a director of the society
was accepted. It was voted to extend
Mr. Marshall a vote of thanks for his
past services to the society.
Mr. J. Arnold Norcross, secretary and
treasurer of the New Haven Gas Light
TRANSACTIONS I. E. S. — PART I
Company, 80 Crown Street, New Haven,
Conn., was elected a director of the
society to fulfill the unexpired term of
the vacancy caused by the election of
Mr. C. O. Bond to the presidency, viz.,
December 12, 1913, to September 30,
191S.
Mr. Alten S. Miller, vice-president of
Humphreys & Miller, Inc., 165 Broad-
way, New York, N. Y., was elected a
director to fulfill the unexpired term of
Mr. Albert Jackson Marshall, resigned,
vis., from December 12, 1913, to Sep-
tember 30, 1 914.
Mention was made of a letter received
from Mr. W. F. Durand, executive sec-
retary of the International Engineering
Congress which is to be held in San
Francisco in 1915. This letter stated
that the Congress is backed by the five
oldest engineering societies and that the
papers to be presented before the con-
gress which might be of interest to
illuminating engineers would be included
with some general papers in a miscella-
neous volume, the contents and char-
acter of which had not yet been deter-
mined.
Invitations to hold the 1914 Conven-
tion of the society in Cleveland were
received from local civic, professional
and commercial organizations and, con-
jointly, from representatives of a num-
ber of manufacturing companies and
several colleges.
The president was directed to appoint
a special committee to consider the time
and place of the 1914 convention. The
aforementioned invitations were referred
to that committee.
It was voted to hold the regular
meetings of the Council during the rest
of the present administration on the
afternoon of the second Thursday of
the month at 2 p. m.
The meeting adjourned at 1 .25 p. m.
Section Notes.
CHICAGO SECTION
The December meeting of the Chicago
Section was held on the 10th at the
residence of Mr. A. D. Curtis of the
National X-Ray Reflector Co. Mr.
Curtis read a paper entitled "Five Years
Progress in the Indirect Lighting of the
Home." A number of different types
of lighting appliances for residence
lighting was shown in connection with
the paper. Seventy-five members and
guests were present.
At the January meeting, which will be
held on the 27th, at the Art Institute,
Mr. J. B. Jackson will read a paper
entitled "Planning Lighting Installa-
tions." Mr. W. A. Durgin will give the
second one of his 20-minute talks on
the "The Fundamentals of Illumination."
NEW ENGLAND SECTION
A meeting of the New England Sec-
tion was held in the Auditorium of the
Edison Illuminating Company of Bos-
ton, December 9, 1913. Mr. L. A. Haw-
kins of the Research Laboratory of the
General Electric Company read a paper
entitled "The Nitrogen-Filled Lamp and
Its Possibilities." One hundred and
forty members and guests were present.
NEW YORK SECTION
A regular meeting of the New York
Section was held in the United Engi-
neering Societies Building, 29 West 39th
Street, New York, December 11, 1913.
Two papers, "Reasons for Styles in
Architecture" and "Periods of Archi-
tecture as Applied to Fixture Design,"
were presented by Messrs. Frank E.
Wallace and Howard E. Watkins, re-
spectively. Both papers called forth
some very interesting discussions. One
hundred and forty members and 40
guests were present. A dinner preced-
TRANSACTIONS I. E. S. — PART I
ing the meeting at Keens' Chop House
in West 36th Street, was attended by
30 members and guests.
PHILADELPHIA SECTION
The Philadelphia Section held a
meeting December 8 at the Engineers
Club, 1317 Spruce Street. The follow-
ing papers were presented : "Railway-
Car Lighting" by G. H. Hulse, "The
Measurement of Brightness and Its Sig-
nificance" by Dr. H. E. Ives, and "The
Mercury Quartz Tube Lamp" by Mr.
M. D. Bucknam. A 220-volt quartz tube
lamp was demonstrated by Mr. H. Cal-
vert. Mr. J. Stilwell exhibited some
water sterilizing apparatus. One hun-
dred and sixty-five members and guests
attended the meeting. At a dinner at
the Engineers Club preceding the meet-
ing 50 members and guests were present.
The program of meetings and papers
for the rest of the season is as follows :
Friday, January 6.
"Deficiencies of the Method of Flicker
for the Photometry of Lights of
Different Colors."
By Prof. C. E. Ferree.
Saturday, February 7.
Meeting under the Auspices of Drexel
Institute.
"Light and How to Use It."
By Mr. C. O. Bond, President
of I. E. S.
Wednesday, February 18.
Joint Meeting with Franklin Institute.
"Artificial Daylight."
By Dr. Herbert E. Ives.
Friday, March 20.
"Lighting and Signalling Systems of
Subways."
By Air. F. D. Bartlett.
"The Sun — The Master Lamp."
By Prof. James Barnes.
Thursday, April 9.
J^int Meeting with Franklin Institute.
"Recent Developments in the Art of
Illumination."
By Mr. Preston S. Millar.
Friday, April 17.
"The Structure of the Normal Eye and
its Ability to Protect Itself Against
Ordinary Light."
By Dr. Wendell Reber.
"Glassware for Illumination and Other
Purposes."
By Mr. James Gillinder.
Friday, May 15.
Mass Meeting of all the Engineering
Societies of Philadelphia and
Vicinity.
Special Program to be arranged and to
include an address on
"The Relation of Engineers to the
Progress of Civilization."
By Dr. Chas. Proteus Steinmetz.
PITTSBURGH SECTION
The November meeting was held on
the 28th in Thaw Hall of the University
of Pittsburgh. Dean F. L. Bishop of
the Engineering School of the latter
university gave a lecture on "The Ptrys-
ics of Lighting." Fifty-three members
and guests were present.
On December 12, the Pittsburgh Sec-
tion held a joint session with the local
chapter of the American Institute of
Electrical Engineers in the Auditorium
of the Engineers Society of Western
Pennsylvania. One hundred and thirty-
two members and guests of both socie-
ties were present. The following papers
were presented : "A Problem in Boule-
vard Lighting" by J. M. Froelich;
"Variables in Street Lighting" by J. F.
Martin; and "Some Aspects of Free
Lamp Renewals" by T. F. Campbell.
The following program of papers and
meetings has been arranged :
TRANSACTIONS I. E. S.— PART I
January i6th.
(i) "A Photographic Analysis of
Diffusing Units with Varying Indirect
Component" by E. B. Rowe and H. H.
Magdsick of the National Electric Lamp
Association, Cleveland, Ohio. The
speakers will have photometric curves
on present types of commercial "semi-
direct" glassware showing the effect
or variations in contour, density and
lamp arrangement on absorption and
transmission, as well as on reflecting
and illuminating efficiency.
(2) "The Relation of the Engineer to
the Problems of Fixture Design" by
A. B. Wilson and F. J. Blaschke of the
National Electric Lamp Association,
Cleveland, Ohio. This paper deals with
the object of a lighting installation, ex-
plains what a fixture really is and illus-
trates its correct and incorrect applica-
tion, mentions co-operation between
engineer and fixture maker, states how
designers' ideas are carried out. It
also shows how the lighting fixture is
an element in the advancement of the
art of illumination.
February 13TH.
"Lighting of Railroad Yards" by A.
C. Cotton of the Pennsylvania Railroad
and Harold Kirschberg of the Lighting
Specialties Company. The authors will
discuss the various elements entering
into the lighting of track scales and
classification yards, together with the
difficulties experienced with same, and
how they are best overcome.
March 13TH.
"Modern Gas Lighting" by S. B.
Stewart, Contract Agent for Consoli-
dated Gas Company, Pittsburgh, Pa.
This meeting will be devoted to the dis-
cussion of gas arcs as applied to modern
illuminating systems. A number of
prominent manufacturers and operators
will be present and take part in the dis-
cussion.
April 17TH.
"The Development of Flame Carbon
Arc Lamps" by C. E. Stephens, West-
inghouse Electric & Mfg. Company.
The author will trace the growth and
development of this popular form of
illuminant from its inception down to the
present time, showing how the difficul-
ties first experienced have been over-
come, and its application to various
fields.
New Members.
At a meeting of the Council held
December 12, 1913, the following 28
applicants were elected members of the
society :
Anderson, C. E.
Educational Department, General
Electric Company, Harrison, N. J.
Arbogast, O. J.
Salesman, Commonwealth Edison
Company, 120 West Adams Street,
Chicago, 111.
Barnitz, Frank R.
Assistant Secretary, Consolidated
Gas Company, 124 East 15th Street,
New York City.
Bertsche, Fred.
Commonwealth Edison Company,
120 West Adams Street, Chicago,
111.
Breuggemeyer, A. H.
Lighting Solicitor, Commonwealth
Edison Company, 120 West Adams
Street, Chicago, 111.
Crowley, Frank M.
Lighting Salesman, Commonwealth
Edison Company, 120 West Adams
Street, Chicago, 111.
Dawson, James.
Salesman, Commonwealth Edison
Company, 120 West Adams Street,
Chicago, 111.
TRANSACTIONS I. E. S. — PART I
Donnelly, Jas. E.
Light Salesman, Commonwealth
Edison Company, 120 West Adams
Street, Chicago, 111.
Griner, Charles A.
Salesman, Commonwealth Edison
Company, 120 West Adams Street,
Chicago, 111.
Hecker, L. M.
Salesman, Contract Department,
Commonwealth Edison Company,
120 West Adams Street, Chicago.
111.
Horan, William H.
Salesman, Commonwealth Edison
Company, 120 West Adams Street,
Chicago, 111.
Hyedahl, Nick.
Commonwealth Edison Company,
120 West Adams Street, Chicago,
111.
Kaeder, F. J.
Salesman, Commonwealth Edison
Company, 120 West Adams Street,
Chicago, 111.
Keys, Harvey A.
Light Salesman, Commonwealth
Edison Company, 120 West Adams
Street, Chicago, 111.
Lancaster, Walter B.
Physician, 101 Newbury Street,
Boston, Mass.
Long, Claude P.
Solicitor, Contract Dept., Common-
wealth Edison Company, 120 West
Adams Street, Chicago, 111.
Lorenz, J. M.
Salesman, Central Electric Com-
pany, 320 5th Avenue, Chicago, 111.
Myers, Romaine W.
Consulting Engineer, 204 Bacon
Building, Oakland, Cal.
O'Brien, John C.
Lighting Salesman, Commonwealth
Edison Company, 120 West Adams
Street, Chicago, 111.
Peck, Edward L.
Inspector, Electrical Testing Labo-
ratories, 80th Street and East End
Avenue, New York City.
Prabel, Frederick C.
Lighting Salesman, Commonwealth
Edison Company, 120 West Adams
Street, Chicago, 111.
Reill, Wilfred J.
Lighting Salesman, Commonwealth
Edison Company, 120 West Adams
Street, Chicago, 111.
Reinach, Hugo B.
Asst. Gen'l. Superintendent, Con-
solidated Gas Company, 124 East
15th Street, New York City.
Rosenberg, Maurice.
General Manager, Shapiro & Aron-
son, 20 Warren Street, New York
City.
Rusch, William T.
Assistant to Engineer of Utilization,
Consolidated Gas Company, 124
East 15th Street, New York City.
Severn, George K.
Salesman, Commonwealth Edison
Company, 120 West Adams Street,
Chicago, 111.
Stilwell, John.
Efficiency Engineer, Consolidated
Gas Company, 124 East 15th Street,
New York City.
Winans, R. K.
Lighting Agent, Commonwealth
Edison Company, 120 West Adams
Street, Chicago, 111.
Additional Sustaining Members.
The following five companies were
elected sustaining members of the
society, December 12, 1913 :
Consolidated Gas Company of New
York.
124 East 15th Street, New York
City.
TRANSACTIONS I. E. S. — PART I
National X-Ray Reflector Company.
235 West Jackson Boulevard, Chi-
cago, 111.
Official Representative : Augustus
D. Curtis.
Public Service Company of Northern
Illinois.
157 South LaSalle Street, Chicago,
111.
Stone & Webster.
147 Milk Street, Boston, Mass.
The Cleveland Electric Illuminating
Company.
Cleveland, Ohio.
Official Representative: R. H.
Wright.
International Electrical Congress.
The International Electrical Congress
is to be held at San Francisco, Septem-
ber 13 to 18, 1915, under the auspices
of the American Institute of Electrical
Engineers by authority of the Inter-
national Electrotechnical Commission,
and during the Panama-Pacific Inter-
national Exposition. Dr. C. P. Stein-
metz has accepted the honorary presi-
dency of the Congress. The delibera-
tions of the Congress will be divided
among twelve sections which will deal
exclusively with electricity and electrical
practise. There will probably be about
250 papers. The first membership invi-
tations will be issued in February or
March, 1914.
Attention is drawn to the distinction
between this Electrical Congress and
the International Engineering Congress
which will be held at San Francisco
during the week immediately following
the Electrical Congress. The Engineer-
ing Congress is supported by the Socie-
ties of Civil, Mechanical and Marine
Engineers and by the Institutes of Min-
ing and Electrical Engineers, as well as
by prominent Pacific Coast engineers
who are actively engaged in organizing
it. This Congress will deal with engi-
neering in a general sense, electrical
engineering subjects being limited to
one of the eleven sections which will
include about twelve papers, treating
more particularly applications of elec-
tricity in engineering work.
The meeting of the International
Electrotechnical Commission will be
held during the week preceding that of
the Electrical Congress.
Nomenclature.
At the next meeting of the Committee
on Nomenclature and Standards, which
will take place in February, the follow-
ing definitions will come up for dis-
cussion :
A direct lighting system is one in
which substantially all the useful light
flux comes directly from the illuminant,
including its diffusing or reflecting
auxiliaries.
An indirect lighting system is one in
which substantially all the useful light
flux is received by diffuse reflection
from the ceiling, walls or other diffusely
reflecting surfaces of relatively large
extent.
Criticisms and comments on these
definitions may be sent to the secretary
of the committee, Dr. Clayton H. Sharp,
80th Street and East End Avenue, New
York, N. Y.
TRANSACTIONS I. E. S. — PART I
FINANCIAL REPORT FOR FISCAL YEAR
ENDING SEPTEMBER 30, 1913.
Wm. J. Struss & Co.
Certified Public Accountants
93-99 Nassau Street
New York
October 24th, 19 13.
Mr. William Cullen Morris.
Chairman, Finance Committee,
Illuminating Engineering Society,
29 West 39th Street,
New York, N. Y.
Dear Sir:—
In accordance with your instructions we have examined the
books and accounts of the Illuminating Engineering Society for
the nine (9) months ended September 30th, 19 13.
The results of this examination are set forth in the two ex-
hibits, attached hereto, as follows :
Exhibit "A "—Balance Sheet, September 30th, 1913.
Exhibit "B" — Earnings and Expenses, for the nine
months ended September 30th, 1913.
We hereby certify that the accompanying Balance Sheet is a
true exhibit of its financial condition as of September 30th, 1913,
and that the attached statement of Earnings and Expenses is
correct.
Respectfully submitted,
Wm. J. Struss & Co.,
Certified Public Accountants.
Illuminating Engineering Society Balance Sheet
September 30TH, 1913.
Exhibit "A"
Assets
Cash on hand and in bank $2,443. 53
Accounts receivable —
191 2 Miscellaneous $ 20.28
1913 Miscellaneous 279.40
1913 Dues 430.00
1913 Initiation fees 120.00
1913 Advertising • 4.46
i9i4Dues-. 120.00
1914 Dues — sustaining members 675.00
1,649.14
TRANSACTIONS I. E. S. — PART I 9
Investments —
Northern Pacific & Great Northern
Railway Bonds — $2,000.00 (book
value) 1,920.00
Furniture and fixtures 894.71
Less depreciation 15 per cent 134.20 760.51
Badges on hand 45-oo
Accrued interest on bonds 20.00
$6,838.18
Liabilities.
Advanced and unearned items —
Advance Dues, 1913 $ 250.00
Fees, 1913 3500
Advertising, 1913 4.93
Dues, 1914 160.00
Fees, 1914 20.00
Dues sustaining members
1914 60.00
Unearned Sustaining members dues 935-42
Dues... • 1,717.50 3,182.85
Outstanding debts (estimated) 1 ,789.93
Surplus Balance, January 2, 1913 1,453-94
Back dues collected 1912 73.00
Convention fund surplus 1912.-. 177-74
Miscellaneous 79-88
Net gain for the nine months,
ended September 30, 1913, as
per Exhibit "B" 80.84 1,865.40
$6,838.18
Illuminating Engineering Society
Statement of Earnings and Expenses for the Nine Months Ended
September 30TH, 1913.
Exhibit "B"
Earnings.
Members dues I5.207.50
Initiation fees 360.00
Advertising 1,097. 14
Profit on badges sold 6.50
Certificates 1.70
Interest on bonds 60.00
Miscellaneous sales of Transactions • • • 211.52
Illumination Primer 616.25
Sustaining members dues 689.58
Annual meeting 6.30
$8,256.49
10 TRANSACTIONS I. E. S. — PART I
Expenses.
Transactions $1,844.72
General office —
Salaries $2,226.14
Rent 633.15
Postage 236.35
Telephone and telegraph • . • • 143 .96
Printing and stationery 294.24
Miscellaneous 235.00 3,768.84
New York Section 245.03
Chicago Section 217.04
New England Section 53 .78
Pittsburgh Section 149-73
Philadelphia Section 130.88
1913 Convention 257.67
1913 Election expense 57-44
Committee on —
Nomenclature and Standards 14-67
Glare 60.00
Research 17-5°
Popular Lectures 4.25
Collegiate Education 10.75
Papers 14.35
Sustaining Membership 6.40
Joint meetings with other
societies 43-57
Depreciation on furniture and
fixtures 134. 20
Miscellaneous expense 100.65
Advance copies and reprints • • 9.93
Exchange and discount 9.25
Unpaid accounts (estimated) —
General office 30.00
Transactions 5.00
Miscellaneous 75 .00
1913 Convention 915.00 1,025.00 8,175.65
Excess of earnings over ex-
penses $ 80. S4
■r-
6'VH
TRANSACTIONS
OF THE
Illuminating
Engineering Society
DECEMBER, 1913
PART II
Papers, Discussions and Reports
[ DECEMBER, 1913 ]
CONTENTS - PART II
The Lighting of Show Windows. By H. B. Wheeler 555
Modern Practise in Street Railway Illumination. By S. G.
Hibben 589
Church Lighting. By Robert B. Ely 613
Some Studies in Accuracy of Photometry. By Evan J.
Edwards and Ward Harrison 633
The Status of the Lighting Art 652
Annual Report of General Secretary. By Joseph D. Israel 683
**■>'•
THE LIGHTING OF SHOW WINDOWS.*
BY H. B. WHEELER.
Synopsis: The lighting of a number of typical show windows is dis-
cussed with respect to the intensity required along the line of trim, and
the selection of reflectors, their spacings and methods of installation.
Illustrations, plans and illuminometer readings and curves of a number
of well lighted windows are given. The material is arranged so as to
make the selection, spacing, and installation of reflectors very simple.
The paper is divided into two sections : The first covers methods of il-
luminating various typical windows. The second gives test data on the
same.
Up to the present time very little data has been presented on
the lighting of show windows. That which has been given
pertains chiefly to the requirements of good show window light-
ing, the influence of surroundings on the treatment of a window
and, occasionally, a description of some one notable installation.
I have assumed that the foregoing matters are well established
facts, and therefore this paper is devoted to methods of obtaining
these good results in window illumination.
Typical show windows have been grouped in three classes
which for convenience are designated as A, B and C; and the
more special windows in two other classes, referred to as D and
E- For each class I show a window of the type under considera-
tion, the reflector which was used, the method of installing the
reflector, and a distribution curve of the reflector.
The questions of selecting the proper reflector for a window,
and the spacings to be used, to give the desired intensity of illum-
ination along the line of trim, are discussed, and I have added
recommendations of the intensities I have found good practise
for show windows. Lamps, shade holders, window drapes (val-
ances), and other features of window trimming that are of vital
interest to the illuminating engineer are discussed briefly. The
paper also includes the results of tests made on typical windows
of the various classes discussed.
* A paper read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
2
556 TRANSACTIONS I. E. S. — PART II
In general there are two types of windows : ( i ) open back ;
(2) boxed-in. In the first class the display is usually very low,
in most cases being practically horizontal. The problem here is
merely to provide the proper intensity of horizontal illumination.
Since these windows are generally quite shallow and high, they
require a concentrating type of reflector for the very shallow
window, and a semi-concentrating for the deeper ones.
The reflectors shown in Figs. 10 and 7, Nos. 755 and 780,
respectively, are most suitable for such windows.
The following classification I believe takes into account all
of the various types of boxed-in windows encountered to-day.
TYPICAL WINDOWS.
Class A Windows : Height equal to depth. High trim. Aver-
age height 9 ft. (2.74 m.). Depth 9 ft. (2.74 m.). Height of
trim 9 ft. (2.74 m.).
Class B Windows: (1) Height 1^ times depth. Medium
trim. Average height 10 ft. (3.04 m.). Depth 6 ft. (1.83 m.).
Height of trim 7 ft. (2.13 m.).
(2) Height 1 J/2 times depth. High trim. Average height
10 ft. (3.04 m.). Depth 6 ft. (1.183 m-)- Height of trim 9 ft.
(2.74 m.)
Class C Windows : Height 2 times depth. Medium trim.
Average height 12 ft. (3.65 m.). Depth 6 ft. (1.83 m.). Height
of trim 7 ft. (2.13 m.).
SPECIAL WINDOWS.
Class D Windows : Height equal to depth. High trim. Aver-
age height 5 ft. (1.52 m.). Depth 6 ft. (1.83 m.). Height of trim
5 ft. (1.52 m.).
Class E Windows : Height 2 times depth. Low trim.
Average height 5 ft. (1.52 m.). Depth 2 ft. 6 in. (0.76 m.).
Height of trim 2 ft. (0.61 m.).
The height of the window is always measured from the floor
to the ceiling; the depth, from the glass front to the back; the
trim, from the floor up.
Class A Windozvs. — In Fig. 1 is shown a Class A window.
whkklkk: the lighting of show windows 557
Windows of this class usually are trimmed up high on the back-
ground, and hence require a reflector which distributes the light
flux over the angle zero to 90 degrees.
The reflector shown in Fig. 2 has been designed to meet this
condition ; its distribution curve is shown in Fig. 4. This re-
flector is a non-symmetrical reflector, with a portion of the
front cut away, to permit the light flux to escape horizontally.
Hence a large portion of the lamp is exposed to view from within
the window. Mirrors should not be placed in the upper part of
the background, in order to avoid the possibility of seeing an
image of the lamp when observing the window from the street.
The practise of using mirrors in any window should be dis-
couraged as images of surrounding objects are generally present,
which detract from the goods on display. One of the chief
reasons why merchants desire mirrors, is the fact that they be-
lieve the observer will be enabled to see both the front and back
of the objects in the window. However, the brightness to which
the back of the objects is illuminated is so low that the results
are not very satisfactory. In addition to this, much brighter
images of surrounding objects serve to detract rather than add
to the effect soueht.
Fig- 1.— A Class A window ; height 9 ft., depth 9 ft., equipped with 60-watt clear
tungsten lamps in reflectors (Fig. 2) on 15-inch centers.
553
TRANSACTIONS I. E. S. — PART II
Fig. 2.— Reflector (No. 777).
Details for installing in
windows shown in Fig. 3.
A/o/e - With unskirted lamps use extension socket,
Fig- 3-— Details for installing in windows reflector shown in Fig. 2.
wheeler: the lighting oe show windows
i59
Z
<
CANDLE. POWER
<x>
z
<
CANDLE POWER
REFLECTOR Ho 777
HOLDER 24 Form '0'
LAMP Clear Tungsten S-2*i -A
WATTS 636 VOLTS »5S
CURVE A
CURVE B
CURVE A
CURVE 8
R
L
R
R
u
R
0
166.3
166-3
166-3
90
151.6
5'
£65.2
110-9
159. S
95
IIO.Z
15
nz.4
385
HZ-S
105
ZS.7
Q
231.0
193
112.7
115
I.Z
35
ZJ1.9
63
34. Z
125
40
135
45
20*. 9
71.7
K5
55
239.7
S8.2
155
65
Z6S.6
39.6
165
75
2413
19.7
175
85
1916
3.4
ISO
Fig. 4-
-Photometric distribution curve and data on reflector No.
Fig. 2, with 1.25-watt-per-candle tungsten lamp.
777, shown in
560
TRANSACTIONS I. £. S. — PART II
u
o
z
<
CANDLE POWER
UI
z
CANDLE POWER
REFLECTOR No-780
HQLPER 314' Form "A"
lamp Clear Tungsten SSO'A
WATTS WOO VOLTS. UO.O
CURVE A
CURVES
CURVE A
CURVE B
R
L
R
R
L
R
0
4/5.0
415.0
4IS.0
90
5
388.0
411-0
408.0
95
15
301.0
Z47.0
408. 0
105
IS
307.0
116.0
400.0
1)5
35
380.0
tt.O
Z97-0
its
40
135
45
3S3.0
4.Z
137.0
W5
55
156.0
246
155
65
BO.S
3.1
165
75
2.4
175
85
185
Fig. 5- — Photometric distribution curve and data on reflector (No. 780) shown
in Fig. 7, with 1.08-watt-per-candle clear tungsten lamp.
wheeler: the lighting of show windows
561
Class B Windows. — In Fig. 6 is shown a Class B window.
The more common windows of this class are trimmed only to a
medium height, but occasionally some are found in which the
trim is carried up high. The first sub-division requires the use
of a reflector which distributes the light flux in the angle zero
to 55 degrees.
The reflector shown in Fig. 7 has been designed to accomplish
this result. Its distribution curve is shown in Fig. 5.
Reflector No. 780 not only confines the light flux where de-
sired, but hides the lamp filament from view within the store,
with even a medium height background. Curtains above the
background are often used to conceal the reflectors, (see Fig.
6.) '
For the second sub-division it is necessary to use a combina-
tion of the reflectors shown in Figs. 2 and 7 in order to get the
desired results.
A solid background is advisable with this latter arrangement
unless suitable curtains are used.
Fig. 6.— A Class B window ; height 12 ft., depth 7 ft., background S ft., equipped
with clear 100-watt tungsten lamps in reflectors (Fig. 7) on 12-iuch centers.
562
TRANSACTIONS I. E. S. — PART II
Fig. 7.— Reflector (No. 780). Details for installing in windows shown in Fig. 8.
Fig. 8.— Details for installing in windows reflector (No. 780) shown in Fig^-7.
wheeler: the lighting oe show windows
563
Class C Windows. — In Fig. 9 is shown a Class C window.
In this class of window the trim is carried up to only a medium
height. A reflector which concentrates the light flux in the
angle intercepted by the line of trim, most nearly meets the con-
ditions. This angle is relatively smaller than for most other
types of windows, and hence presents a much harder problem to
be dealt with. It is not only necessary to get high concentration
on the floor of the window, but the background also must be
properly illuminated to the top of the trim, and the light flux
cut-off quite sharply at this point.
Fig. 10 shows a reflector which has been designed to give the
required results. Its light distribution curve is shown in Fig. 12
Fi&- 9-— A Class C window, height 16 ft. 6 in., depth 8 ft., background 8 ft., equipped
with ioo- watt clear tungsten lamps in reflectqrs (Fig. io) on i8-inch centers.
564
TRANSACTIONS I. E. S. PART II
Fig. io.— Reflector (No. 755) details for installing in windows shown in Fig. 11.
Fig. 11. — Details for installing in windows reflector (No. 755) shown in Fig. 10.
WHEELER : THE LIGHTING OF SHOW WINDOWS 565
IP
z
CANDLE POWER
Ui
_i
z
<
CANDLE POWER
1
REFLECTOR 4* 7&
holder ,# Form'A
L^MP Tungsten S-JO-AC/ecr
WATTS 1000 VOLTS llOO
CURVE A
CURVE 8
CURVE A
CURVE B
R
L
R
R
L
R
0
926.0
9Z6.0
926.0
<»
5
895.0
677.0
931.0
95
15
617.0
66S.0
7S7.0
105
15
497.0
z».o
fd&O
115
35
mu>
70.0
130,0
IZS
40
135
45
2800
6-6
S.'.B
MS
55
IIZ.0
to
164
155
65
liS
7.1
165
75
17
3J
175
85
180
Fig. 12.— Photometric distribution curve and data on reflector (No. 755)
shown in Fig. 10, with a 1.08-watt-per-candle lamp.
566
TRANSACTIONS I. E. S. — PART II
1
CANDLE" POWER
ex.
CANDLE POWER
REFLECTOR M- 7S0
HOLDER Special attached
LAMP Clear Tungsten J- Zl- E
WATTS IZO-0 VOLTS l'0-O
CURVE A
CURVEB
CURVE A
CURVE B
R
L
R
R
L
R
0
4ZBM
428.0
4ZS.0
90
s
fJZ.O
413.0
t28.0
95
15
448.0
330J/
404,0
105
25
430M
ZZ3.0
354.0
115
J5
401.0
131.0
zaz.o
125
40
135
45
356.0
6tt
116.0
145
55
314.0
Z4.0
116.0
155
65
Z44.0
6ZJ)
165
75
96.6
3&S
175
es
IZ.I
1S.7
180
-Photometric distribution curve and data on reflector (No. 750) shown
in Fig. 15, with 1.12-watt-per-candle clear tungsten lamp.
wheeler: the lighting of show windows 567
Class D Windozvs. — This class of window is found chiefly in
cases where the show window is divided into two tiers, and in
shops with low head room, situated on the ground floor.
The light distribution required is essentially that of the reflector
shown in Fig. 2, but there is not sufficient head room in these
windows to permit its use.
Fig. 15 shows a reflector which has been designed to meet the
latter condition. Its light distribution curve is shown in Fig.
13. This reflector is provided with an adjustable holder as
shown in Fig. 16, and the lamps are placed horizontally in the
reflector. It is usually installed with the flat side tilted at an
angle of approximately 15 degrees with the horizontal. A mir-
ror background should not be used in a window of this type for
obvious reasons.
Class E Windozvs. — In Fig. 18 is shown a class E window.
This class of window is found largely in jewelry stores, cigar
stores, and shoe stores. The line of trim is low, and frequently
it is practically flat. This window requires a reflector having
a light distribution curve similar to that of the reflector shown in
Fig. 10. This reflector of course is too large for a window of
this size.
The reflector (Fig. 19) installed as shown in Fig. 20, has been
largely used for this type of window. Its distribution curve is
shown in Fig. 14. A window may be illuminated in this man-
ner very satisfactorily, but since the reflector is symmetrical it is
not so economical, inasmuch as a great deal of the light flux es-
capes to the street and upper portion of the window. The
percentage of the total light flux incident on the surface of the
line of trim is low.
At the present time a small non-symmetrical reflector like that
shown in Fig. 10 is being developed for this class of windows.
Preliminary tests on this new reflector with an unskirted base
60-watt clear bulb tungsten lamp show a highly concentrated
light distribution very similar to that of Fig. 12. The cut-off
at the window side of the reflector is slightly sharper than for
the No. 755 reflector.
568
TRANSACTIONS I. E. S. — PART II
4*1
jr
i.
CANDLE POWER
<
CANDLE POWER
LAMP AND REFLECTOR
BARE LAMP
LAMP
&RER-
BARE
LAMP
LAMP
SrREFL.
BARE
LAMP
ZONE"
0-60
0-90
0-190
0-60.
0-90
0-80
LUMENS
Wft?
£Z£7
■i"^?
SZf.O
0
306.0
90
REFLECTOR M.m H0LPER#'^/ot 0
LAMP Tungsten WATTS 6°-° vo\js-M-0
5
ViSJ)
95
IS
234.0
105
IS
175.0
115
35
MO.O
\15
40
135
45
108.0
145
S5
S6.6
155
65
19.0
165
75
175
85
\B0
Fig. 14. — Photometric distribution curve and data on reflector shown
in Fig. 19, with a 1.12-watt-per-caudle lamp.
wheeler: the lighting of show windows
569
Fig. 15.— Reflector (No. 750). Details for installing
in window shown in Fig. 16.
I-Condulef.
Z-Pubber Gasket.
3- Receptacle.
Plug
Section and Plan
showing installation
de tolls of
Reflector No. 750
I 40 Walt Clear Bulb
J Lamp
Fig. 16. —Details for installing in windows reflector
(No. 750) shown in Fig. 15.
57o
TRANSACTIONS I. E. S. — PART II
Fig. 17. — Test window set up as a Class B window.
Fig. 18. — A Class E window ; height 5 ft., 3 in., depth 3 ft., equipped with 60-watt clear
tungsten lanips in reflectors (Fig. 19) on 15-inch centers.
wheeeer: the lighting oe show windows 571
Fig. 19.— Reflector (No. 696). Details for installing
in windows shown in Fig. 20.
60 Watt Cleat- Bulb
Reflector A/o 696. Lamp
(Unskirted base. )
Fig. 20.— Details for installing in windows reflector (No. 696) shown in Fig. 19.
Fig. 21. — Show window drape (valance).
572
TRANSACTIONS I. E. S. — PART II
CEILING LINEv
u /
<
I4\
14
r
13 \
13
»2\
12
"\
II
UJ
UJ
®
?.\
8
7
6
10
. -*^-
_S^.
X-
^
k..-^S
(c>N
^-
-— ^
8
Q
g
UJ
1
1
1/ o
1 <
1 u.
\ o
■fc
is
x
to
•z.
6
u.
o
o
•z.
5
4-
5
\ 4
3
o
o
2
2
X
-NO. 780
-NO. 755
u?
1
1
UJ
DC
^V^"
i
2
3
4
5
6
7
8
s i<
UJ >
CO /
FLOORS (§)
!
Fig. 22. — Chart for the selection of the proper reflector.
wheeler: the lighting oe show windows 573
SELECTION OF REFLECTORS.
The chart shown in Fig. 22 makes easy the selection of the
correct reflector for any type of window. To start with, knowl-
edge of three things is necessary : height, depth of window, and
height of trim at background.
Suppose, for example, the show window is 10 ft. (3.04 m.)
high, 6 ft. (1.83 m.) deep and the trim or background to be
lighted is 6 ft. (1.83 m.) high. The procedure would be as
follows :
First, find the height of the window (10 ft.) on the left-hand
vertical scale. This point is indicated at "A."
Second, locate depth of window (6 ft.) on lower horizontal
scale. This point is indicated at "B."
Third, straight up from the latter point locate another point
corresponding to the highest point to which the window is
trimmed (in this case 6 ft.). This point is indicated at "C."
Next note the diagonal line that most nearly passes through
the two points, "A" and "C" (which in this case is a heavy dark
line). By referring to the key below one finds the reflector
designated by the heavy dark line — — — ■— — ^ — — ^—
is reflector No. 780, which is the correct reflector for this win-
dow. In the above example, if the window were 12 ft. high, the
chart would call for reflector No. 755. This chart is based on
the use of the lamp for which each reflector herein mentioned is
designed.
Where it is necessary to place reflectors on the transom bar,
they may be selected by using the distance from the floor of the
window to the bar as the height of ceiling.
SPACING OF REFLECTORS.*
For Class A-B-C windows the spacing, or distance from
center to center for the respective reflectors, is about as follows :
* For Class D and E sec tables 7 and 8 respectively for average conditions.
574
TRANSACTIONS I. I
J. S. — PART
II
Population of
town or city
TABLE I.— Refxector Spacings.
Size of
Reflector Spacing lamp watts
Average
foot-candles
along
line of trim
10,000 and under
No. 777 (Fig. 2)
28 in.
60
10
No. 780 (Fig. 7)
36 in.
100
10
No. 755 (Fig. 10)
48 in.
100
10
10,000-30,000
No. 777 (Fig. 2)
18 in.
60
15
No. 780 (Fig. 7)
24 in.
100
15
No. 755 (Fig. 10)
36 in
100
15
30,000 and up
No. 777 (Fig. 2)
10 in.
60
30
No. 780 (Fig. 7)
12 in.
100
30
No. 755 (Fig. 10)
18 in.
100
30
The above table gives average illumination intensities found
good practise in the live business centers of a town. For stores
on the outskirts and in the outlying districts, reduced intensities
will often be ample. Also due allowance must be made when the
interior finish and trim is darker or lighter than the light oak
for which the table has been compiled.
LAMPS.
The window reflectors discussed in this paper are designed
exclusively for use with the standard tungsten lamps. Each
reflector is designed for a certain size of lamp. All test data
included in this paper is based on the use of the proper sizes of
lamps for which the reflectors are designed.
REFLECTOR HOLDERS.
Owing to the fact that the many makes of shade holders
differ materially in dimensions, I have found it essential to meas-
ure all known makes and determine which of these insure the
proper position of lamp with respect to the reflector. It is quite
essential that this position should be correct, if the best results
are desired.
Fig. 23 shows a portion of a lamp, reflector, and holder. The
values of dimensions F and G for the various reflectors are tabu-
lated below, and the trade names of the shade holders having
these dimensions are given.
If all shade holders were standardized it would be of infinite
value to all concerned.
wheeler: the lighting of show windows
575
TAB
LE II.
Porcelain
Brass
F
G
receptacles
shell sockets
2% ins.
i in.
A-C
B-H-X-S-P
2% ins.
I 15/16 in.
A-C
B-H-P
1% ins.
1 15/16 in.
A
H
2% ins.
1 in.
A-C
B-H-X-S-P
Appleton ;
C--Crouse-Hinds ; H— Holophane
Reflector Holder position
Fig. 2 Form "O"
Fig. 7 " "H"
Fig. 10 " "A"
Fig. 19 " "O"
Symbols used : A —
B— Bryant ; P— Plume & Atwood ; X-Hubbel ; S— P. & S.
Fig. 23.— Diagram of reflector and lamp socket.
WINDOW DRAPES (VALANCES).
In order to conceal show window reflectors from the view of
the observer on the street, window drapes or valances are very
much used. In addition to performing the above functions, they
add a touch of refinement and exclusiveness to a show window,
greatly increasing the attractiveness of the merchandise dis-
played. Many attractive designs are now being manufactured
for this use. Fig. 21 shows one type in which the firm's mono-
gram is inscribed.
TEST METHODS AND DATA.
Test Windows. — Experimental windows for use in testing the
illumination of Class A, B, C, D and E windows, were made
from wood frames of various sizes, over which was stretched a
heavy paper backed with cloth, in much the fashion of stage
scenery. This paper was the color of natural light oak. By
using various combinations of the frames at hand, the typical
windows were readily constructed. For the floors the same type
of paper in a darker oak was used.
Fig. 17 shows the test window set up as a class B window,
with reflectors like that of Fig. 7 in place. From the illustration
it will be noted that the reflectors and lamps are supported from
the 2 in. x 4 in. (5.08 cm. x 10.16 cm.) strip, which was supported
576 TRANSACTIONS I. E. S. — PART II
from two standards having raising and lowering features. Pull
sockets on 6-inch (15.24 cm.) centers the entire length of the
wood strip, made it possible to obtain the various reflector spac-
ings desired.
A black curtain was hung at the front of the window in the
position of the glass in an actual window. In windows having
a medium height of trim in which the upper portion of the
background is glass, as for instance Class B, a black curtain was
hung in this position. The black cloth which reflects a negligible
amount of light flux, was considered a good substitution for
clear glass. Whatever slight difference this substitution may
make, I feel it is of small moment and that the results obtained
are representative of the various classes of windows tested.
The line of trim for the various classes of windows was chosen
for the condition I found existing in the majority of windows.
Instruments. — The instruments used for this series of tests
were a Sharp-Millar portable photometer with a standard lamp
and detached test plate of white diffusing glass mounted on black
velvet; a Weston milli-ammeter with shunt and shunt leads; and
a Wagner A. C. or D. C. voltmeter. All instruments were care-
fully calibrated, both before and after tests. The mean hori-
zontal candle-power and wattage of all lamps used, were deter-
mined before beginning the tests. The milli-ammeter was used
to measure the current passing through the standard lamp. The
volt meter was used to measure the voltage at the lamp terminals.
Method of Conducting Tests. — Tests of the normal illumina-
tion along the line of trim were made at the stations as indicated
for each class of window in the subsequent figures. The loca-'
tion of the stations was accomplished by measuring the hori-
zontal distances from front to rear of window, and marking the
vertical distances on the adjustable standard used for supporting
the detached test plate.
The photometer was placed in such a position that the line of
sight of the instrument would be practically normal to the line of
trim at each station. The reason for this was to facilitate plac-
ing the test plate parallel with the line of trim at any one station,
since it was found that with a little practise the operator was able
to locate by eye the test plate normal to the line of sight of the in-
wheeler: the lighting oe show windows 577
strument. By experiment it was also found that it is not necessary
to view the test plate normally. In fact readings when viewing
the test plate at angles of 45 degrees either side of the normal,
show but slight difference from the normal reading. Hence,
readings of normal illumination along the line of trim taken
with the photometer at the ordinary height of the eye, would not
materially differ from those presented.
Readings of the voltage at the lamp terminals were made
simultaneously with the illumination intensity readings.
In order to obtain some idea as to the relative co-efficient of
reflection of the walls, ceiling and floor of the test windows, the
following test was conducted :
The test plate used throughout the test, was set up in a given
position, under the illumination from tungsten lamps, and the
intensity of illumination incident thereon read.
The test plate was replaced by a disk of white blotting paper
and another reading taken with the photometer. The white
blotting paper was in turn replaced by a sample of the paper used
on the walls and floor of the test window.
Assuming a reflection co-efficient of 100 per cent, for the test
plate, the reflection co-efficient of the blotting paper and walls
and floor of the test window, in terms of the white test plate was
determined.
The values obtained are as follows :
1 : Test plate 100 per cent.
2 : Blotting paper 100 per cent.
3 : Walls of test window 58.4 per cent.
4 : Floor of test window 34.6 per cent.
Later the absolute co-efficient of reflection of the above blott-
ing paper1 was found to be 75 per cent.2 From this it follows
that the absolute co-efficient of reflection of the above surfaces
are approximately (limited by the accuracy of the test) as
follows :
Per cent.
test plate 75.0
Blotting paper (see Notes below) 75.0
Floor of test window 25.9
Walls of test window 43.8
1 140 Star.
2 Determined by Mr. M. ^uckiesh, of National Electric I^anip Association.
57§ TRANSACTIONS I. E. S. — PART II
These values of course are not the absolute reflection co-effi-
cients of the surfaces, but may be used comparatively, inasmuch
as it is known that white blotting paper reflects about 75 to 82
per cent, light.
Method of Calculating Test Results. — In calculating the test
results of illumination, foot-candle readings were corrected for
115 volts, the rated high voltage of the lamps used. These cor-
rections were made in accordance with the lamp data supplied
by the lamp manufacturer.
The effective lumens on the surface of the line of trim were
calculated from the average of all the foot-candle readings taken
in any one window. The actual wattage and mean horizontal
candle-power of all lamps being known for 115 volts, the effec-
tive lumens per watt and the efficiency of utilization were calcu-
lated. This data is presented in the illumination tables for each
class of window tested.
A floor plan and an elevation at one of the planes tested, is
shown with the data on each of the windows. On the floor plan
are indicated the location of the test planes and the location, num-
ber and spacing of the outlets. On the elevation are shown the
line of trim, the test stations, and the reflector showing its po-
sition relative to the line of trim.
In addition to this a curve showing the normal illumination
at the various test stations is given. The method of plotting
this curve is rather peculiar, inasmuch as it is neither a polar nor
a rectangular co-ordinate curve. The line of trim is indicated
as the zero line of illumination intensity, and the various lines
parallel to the line of trim are given foot-candle values as indi-
cated. The length of the ordinate from the line of trim to the
curve, represents the foot candle intensity of normal illumination
at the station.
This curve drawn as it is, I believe is contrary to all precedent,
and while it cannot be -used for obtaining close approximations
of the illumination intensity at points between stations, I believe
it is a good method for giving one a graphical idea of the distri-
bution of illumination intensity along the line of trim.
wheeler: the lighting of show windows
579
Fig. 24. — Elevation through Class A window
showing uniformity of normal illumina-
tion along the line of trim.
Fig. 25.— Plan of Class A window showing
location of test planes and ceiling outlets.
TABLE 3.— Showing Test Re-
sults in Class A Window
with Reflector No. 777
Shown in Fig. 2.
Centers, 18 in.
Lamps, 60- watt, tungsten clear
skirted.
Total watts, 501.9.
Number of units, 8.
Holder, 2% in. "O."
Watts per running foot, 41.82.
Average
foot-
candles
Plane
A....
Station
. .. I
2
3
4
5
6
7
8
9
. • 1
2
3
12.4
16.0
19.2
16.3
17.7
18.5
23-7
22.7
17.4
10.0
13-9
16.6
Plane
B....
Station
... 4
5
6
7
8
9
. . . 1
Average
foot-
candles
12.3
14.9
17.I
19.8
19-3
14.2
13-7
16.3
15.9
13-4
11. 6
12.2
14.0
13.S
1 1 -5
Average foot-candles, 15.75
Area of surface along line of
trim. 168 sq. ft.
Effective lumens, 2650.
Effective lumens per watt, 5.27.
Total lumens of lamps alone, 4533.
Efficiency of utilization, 58.3 per
cent.
58o
TRANSACTIONS I. E. S. PART II
Fig. 26. — Elevation through Class B window
(medium trim) showing uniformity of
normal illumination along line of trim.
TABLE 4.— Showing Test Re-
sults in Class B Window
( Medium Trim ) with Reflec-
tor No. 780, Shown in Fig. 7.
Centers, 24 in.
Lamps, 100-watt clear tungsten.
Total watts, 501.5.
Number of units, 5.
Holder, 3^ in. "A."
Watts per running foot, 50.15.
Plane
Station
Average
foot-
candies
A
24.4
2
21.7
3
23.I
4
2I-3
5
19-5
6
.10.0
B
7
2.3
20.1
21.7
2
3
24.2
PIRN
1_
Fig. 27. — Plan of Class B window (medium
trim,) showing location of test planes
and ceiling outlets.
Plane
Station
Axerage
foot-
candles
B
4
18.8
5
22.3
6
II. I
7
2.8
c
17.8
17.7
2
3
17.0
4
i5-i
5
15-2
6
11. 8
7
3-o
Average foot-candles, 16.22.
Area of surface along line of
trim, 95.0 sq. ft.
Effective lumens, 1540.
Effective lumens per watt, 3.08.
Total lumens of lamps alone, 4745.
Efficiency of utilization, 32.5 per
cent.
y*x
\ei~-/7£ "A- /4i "■+> aj '^> 10 i
~«
6
O
'
-0"
S"
>-
PLAIN. %
r
sNO. 780 REFLECTOR %
& WITH I0OW. LAMP £
-^- &N0. 777 REFLECTOR &
WITH 60 W. LAMP §
s
4
PLANS A'
O
<V
+
PLANE ft
i
4
♦ 1
QO PLANE 8
t )
PLANE C
5
,T-?-
4i
T
1 I
•
Fig. 28. — Elevatioti through Class B window
(high trim) showing uniformity of nor-
mal illumination along the line of trim.
TABLE 5.— Showing Test Re-
sults in Class B Window
(High Trim) with a Com-
bination of Reflectors Nos.
780 and 777 Shown in Figs. 7
and 2 Respectively.
Centers, 18 in.
Lamps, 4 100-watt clear tungsten
skirted.
Lamps, 4 60-watt clear tungsten
skirted.
Total watts, 653.55.
Number of units, 4 No. 7S0.
Number of units, 4 No. 777.
Holder, No. 780, 3^ in. "A."
Holder, No. 777, 2% in. " O."
Watts per running foot, 54.46.
Average
Plane Station foot-candles
A 1 21.9
Plane
Al .
B
A 1
24.2
22.6
27.7
22.5
18.3
18.3
17.9
20.5
28.1
24.8
21.0
Fig. 29. — Plan of Class B window (high
trim) showing location of test planes
and ceiling outlets.
Average
foot-
candles
20.3
15-6
16.5
23.2
20.3
22.9
23.2
21.0
18.9
16.3
13-6
14.8
19.10
21.3
22.1
21.8
21.4
16.2
151
8 10.4
Average foot-candles, 20.23.
Area of surface along line of
trim, 120.00.
Effective lumens, 2430.
Effective lumens per watt, 3.73.
Total lumens of lamps alone,
6023.
Efficiency of utilization, 40.3 per
cent.
Station
•• 5
6
7
8
.. 1
2
3
4
5
6
7
8
. . 1
2
3
4
5
6
7
3
582
TRANSACTIONS I. E. S. — PART II
[6i~/6£^k-/3"X.
Fig. 30. — Elevation of Class C window show-
ing uniformity of normal illumination
along the line of trim.
TABLE 6.— Showing Test Re-
sults in Class C Window
with Reflector No. 755
Shown in Fig. 10.
Centers, 36 in.
Lamps, 100-watt tungsten clear.
Total watts, ^409.6. J
Number of units, 4.
Holder, $% in. "O."
Watts per running foot, 34.13.
Plane
Station
Average
foot-
candles
21.8
2
22.2
3
18.4
4
14.6
5
12.6
6
7-7
7
4.1
B
19.7
2
22.9
PL /J A/
<K
PLANE A.
PLANE 8.
1 t i I
PLANE C
Fig. 31. — Plan of Class C window showing
location of test planes and ceiling
outlets.
Average
foot-
Plane
Station
candles
B
21.8
4
l6.2
5
u-3
6
7-3
7
4.1
c
1
16.5
2
18.4
3
18.0
4
15-4
5
12.3
6
7-9
7
4-3
Average foot-candles, 14.16.
Area of surface along line of
trim, 114 sq. ft.
Effective lumens, 1610.
Effective lumens per watt, 3.93.
Total lumens of lamps alone,
3862.
Efficiency of utilization, 41.7.
wheeler: the lighting of show windows
583
!>
NO.750 REFLECTOR JOOT^
l>6t*- /8"4- 174 "X* i4i^» n'Xsk
Fig. 32. — Elevation of Class D window show-
ing uniformity of normal illumination
along the line of trim.
Fig- 33— Plan of Class D window show-
ing location of test planes and ceiling
outlets.
TABLE 7.— Showing Test Re-
sults in Class D Window
with Reflector No. 750
Shown in Fig. 15.
Centers, 24 in.
Lamps, 60-watt tungsten clear
short base.
Total watts, 623.4.
Number of units, 5.
Holder, special.
Watts per running foot 62.34.
Plane
Station
Average
foot-
candles
A
34-2
2
37-6
3
45-o
4
42.8
5
37-2
6
26.0
B
38-6
2
43-6
Plane
Station
Average
foot-
candles
B
3
4
47-5
46.7
c
5
6
35-5
25.8
26.9
38.6
2
3
43- 0
4
37-3
5
6
33-4
22.7
Average foot-candles, 36.8.
Area of surface along line of
trim, "86. 24 sq. ft.
Effective lumens, 3170.
Effective lumens per watt, 5.08.
Total lumens of lamps alone,
5642.
Efficiency of utilization, 56.2 per
cent.
584
TRANSACTIONS I. E. S. — PART II
Fig. 34.— Elevation of Class
E window showing uni-
formity of normal illum-
ination along the line of
trim.
TABLE 8.— Showing Test Re-
sults in Ci,ass E Window
with no. 696 reflectors,
Shown in Fig. 19.
Centers, 24 in.
Lamps, 60-watt tungsten clear
short base.
Total watts, 372.45.
Number of units, 6.
Holders, 2% in. " O."
Watts per running foot, 31.04.
Plane
Station
Average
foot-
candles
A
30.0
2
32.9
3
28.3
4
28.O
5
20.3
1
25.5
[—
Z:6
■^ *
i
|
PLPN. §
Si
As G
$~
PL ONE fl
O
+
PLPNE B.
V
CO
t"
"*■
»
<0
*
M-
J
PLANE C
<\\
5
\
<P
^
|
t
\
"
Fig. 35. — Plan of Class
E window showing
location of test
planes aud ceiling
outlets.
Plane
Station
foot-
candles
B
30.I
3
26.2
4
30-5
5
19-3
c
18.6
23.6
2
3
21.7
4
21.8
5
16.2
Average foot-candles, 24.8.
Area of surface along line of
trim, 40 sq. ft.
Effective lumens, 992.
Effective lumens per watt, 2.67.
Total lumens of lamps alone,
3349-
Efficiency of utilization, 29.7 per
cent.
wheeler: the lighting of show windows 585
The results obtained in Class B and C windows I believe are
best. A high intensity of illumination at the bottom and front
of the window with a gradual decrease up to the top of trim
seems to give the desired stage light effect in this class of win-
dows.
In the Class A, D and E windows the illumination intensity
along the line of trim is nearly constant. A very good effect is
secured in each case.
Tests in the various windows run with the reflectors with var-
ious spacings, as would be expected, showed that the average
intensity of illumination along the line of trim varies practically
inversely with the spacing of the units.
The tests in Class A, B and C windows reported here, are
representative of the intensities I have recommended for the
average size city. Windows of the classes D and E are more
prevalent in larger cities and hence the tests reported here show
higher intensities of illumination.
The chart for determining the type of reflector required for
a given window, and the table giving the spacings for the various
reflectors, check out with the test results.
The reason that the No. JJJ and No. 750 (Figs. 2 and 15
respectively) reflectors show efficiencies of utilization consider-
ably above any of the other reflectors, is evident because the
area in which the light is to be distributed is large and takes
in a wide angle. No difficulty is experienced in confining the
maximum flux from the lamp in this area.
In the case of the No. 755, No. 780 and No. 696 reflectors,
(Figs. 10, 7 and 19 respectively), the area in which the light
flux is to be directed is relatively smaller, and takes in a much
smaller angle. It is this fact that causes the resultant lower
efficiencies of utilization. Of course it would not be desirable
to cut off all light outside the angle subtended by the line of
trim, but for the best effect and most efficient results the largest
portion of the light flux must be confined to this angle.
The No. 696 reflector (Fig. 19) as has been previously ex-
plained, allows too much light to escape outside of the angle sub-
tended by the line of trim. A class E window equipped with a
586 TRANSACTIONS I. E. S. — PART II
reflector having a distribution of light like that of Fig. 10, would
show a much better efficiency of utilization.
The reflector shown in Fig. 7 should not show a lower efficiency
of utilization than the reflector of Fig. 10. The difference shown
by the test is due to the fact, that the No. 780 (Fig. 7) reflectors
tested, happened to be the first ones taken from the mold.
This mold after a few reflectors have been run, becomes much
smoother and the later reflectors show a very much increased
efficiency. The efficiencies of the latter two reflectors used for
these tests have been determined and show approximately the
same difference.
In order to obtain an idea as to how much the walls and ceiling
of the windows add to the total illumination on the line of trim
by reflection, tests were run on a class A window with the walls,
ceiling and floor covered with black cloth. The conditions of the
previous test in the class A window were duplicated with the
exceptions noted above. The result of these tests were as
follows :
Average foot-candles 1470
Area of surface along line of trim 168 sq. ft.
Effective lumens 2470
Effective lumens per watt 4.92
Total lumens of lamps alone 4546
Efficiency of utilization 54.3 per cent.
Comparing these figures with those on the light oak window,
it is apparent that 7.37 per cent, light flux is reflected onto the
line of trim by the walls and ceilings of the window. This added
illumination will be more or less depending on the type of window
and interior finish. The light oak window was chosen as a good
average finish.
CONCLUSION.
As noted at the beginning of my paper, the subject of win-
dow lighting has received only a limited amount of attention
from the society, and I trust I have started something which will
lead to more investigations in this direction. It is a subject well
worthy of the attention of the illuminating engineer, the re-
flector manufacturer, the lamp manufacturer, the electrical con-
tractor, and the central station. It is always possible to interest
wheeler: the lighting of show windows 587
a merchant in the true advertising value of a well illuminated
window display, because he can see increased business. It is a
monetary consideration with him. He is open to conviction
when the question of increasing sales is under consideration.
In addition to this, however, beautifully illuminated show win-
dows enhance the beauty of a city, and thus will receive the
endorsement of the populace as a public benefit.
I wish to acknowledge valuable suggestions in preparing this
paper from Messrs. J. R. Cravath, J. B. Jackson, and L. V.
James, associate, electrical engineering department, University
of Illinois, and for the assistance of the engineering department
of the National X-Ray Reflector Co.
Note : The reflectors referred to by number in this paper are manufactured by the
National X-Ray Reflector Company, Chicago, 111.
DISCUSSION.
Mr. J. R. Cravath, Chicago : I believe with Mr. Little that
the surface brightness in the line of trim as seen from the street
should be the criterion by which we should make our installation,
but I also think that Mr. Wheeler's method practically gives us
that within a very narrow margin of error, because we can
see by his method that he moved a diffusing test plate up and
down almost along the line of trim or very close to it.
The main point I want to make is something in regard to
architecture of show windows. It has been said that we must
take show windows as we find them ; but there are frequently
cases where a merchant is remodelling his show windows to
make them more effective and the illuminating engineer ought
to be in a position to recommend the dimensions which will
make it possible to bring out the goods to the best advantage
We all know that show windows are at their best at night under
artificial illumination rather than in the day time; therefore
their proportions should be selected or should be designed to
give the best results under artificial -illumination. Experience
has proven that the type of window which Mr. Wheeler classi-
fies as a Class C, that is, one with the height two times the depth,
does not give good results with artificial light ; the result simply be-
ing that the vertical illumination that shines on the goods is com-
paratively low. This applies to dry goods. Of course with jew-
588 TRANSACTIONS I. E. S. — PART II
elry it is altogether different. It is almost impossible to avoid
having dry goods cast shadows on themselves in such shallow
windows. I prefer for dry goods a deep window as that type
makes possible the best illumination of the goods.
Mr. W. F. Little) : In addition to this very valuable paper,
would it not be well to make further investigations showing
the surface brightness of the various objects in the line of trim?
While the foot-candle intensities as given by Mr. Wheeler are
of prime importance, still surface brightness measurements
showing the range of contrasts met with in a window trimmed
with various materials would be of considerable interest.
Mr. H. B. WhEEEER (In reply) : Regardless of the class a
window falls into, the practise of using mirrored backgrounds
should be discouraged, as images of surrounding articles are gen-
erally present, which detract from the goods on display.
Under typical windows in the large cities on the main streets,
the merchant is very anxious to obtain the full advertising value
of his window ; and hence demands a high degree of illumination.
Thirty-foot candles has been found ample illumination for typical
windows with medium decorations.
From the data presented in this paper, you will note that con-
centrating reflectors are used in comparatively high windows,
and hence are at a considerable distance from the goods on dis-
play. Further the reflectors are designed to allow plenty of
radiation. I have never heard of any case where the light rays
from commercial types of concentrating reflectors, had ever done
any damage in the least to the most delicate goods in a window.
A further investigation of surface brightness would be very
interesting, and I trust that further investigation on this subject
will be gone into.
My paper was based on average conditions typical of fairly
modern show windows. Medium colors were selected as fairly
representative for the ends, background and ceiling of the
various windows tested.
HIBBEN : STREET RAILWAY ILLUMINATION 589
MODERN PRACTISE IN STREET RAILWAY
ILLUMINATION.*
BY S. G. HIBBEN.
Synopsis: Up to the present time street car lighting has been done
inefficiently, with bare carbon filament lamps. Recently four standard
tungsten lamps have been placed on the market, which in conjunction
with proper downward reflecting shades have enabled the energy cost for
lighting to be cut in half, at the same time allowing an increase in the
usable light of more than 80 per cent. Special fixtures are available for
supporting the shades, and selector switches may be employed to insure
continuous lighting service of series burning lamps. Steadier illumina-
tion is secured by the new system of lighting, under the adverse condi-
tions of voltage fluctuations, and the glare that exists at present in the
majority of electrically lighted street cars is done away with through
the medium of the shades. The scientific, economical lighting of street
cars is a field that is full of interesting possibilities, and is now being
very rapidly developed.
Although the incandescent filament lamp has been used for
the lighting of street railway cars practically ever since the elec-
tric motor cars superseded the horse or cable cars, it has not
been until within the last six years that any attempts have been
made to utilize the generated light to best advantage by means
of scientifically manufactured shades or reflectors ; and it has not
been until about one year ago that the tungsten-filament lamps of
an efficiency of 1.4 watts per horizontal candle or better, have
been perfected to the extent of being sufficiently rugged for
this street railway service. Consequently the economic and scien-
tific illumination of street cars has been slow in its inception and
development, compared with the rapid progress that has been
made in the lighting of large office buildings, stores, residences,
or even steam railway cars.
There are several reasons for this slow development. Lighting
energy, being but a small fraction of the total energy used by
motors, and being relatively cheap to generate, has not been
considered as a fit subject for economy. Also the short periods
* A paper read at the seventh annual convention of the Illuminating Engineering
Society. Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
590 TRANSACTIONS I. E. S. — PART II
that individuals of the traveling public use the lighting, and the
disposition of passengers to make the best of the illumination,
unless it be absolutely unbearable, has not been conducive to
public protestations. Furthermore the rough usage to which
lamps and shades are necessarily subjected, the low first cost
of the carbon filament lamps, the uncertain relations between
private street railway corporations and municipalities, have all
been reasons for slowness in arriving at modern car lighting.
How really vital these reasons are will be left to be judged after
a further consideration of this paper.
As far as the author can determine, the first street car using
individual reflectors on lamps was put in service in 1909, this
being a car of the Oakwood Traction Co., operating in Dayton,
Ohio, although previously there may have been a few desultory
attempts to equip lamp clusters with reflecting glassware. This
car was equipped with center-deck 4-light fixtures, and side wall
single light brackets, using square shaped alba glass shades. A
view of this car is shown in Fig. 1. Cars with this equipment
are still in service.
About two years ago a number of traction companies installed
bare 23 and 36-watt tungsten filament lamps replacing the
carbon lamps, having their attention directed chiefly towards de-
termining the ultimate lamp life. When the majority of these
new lamps had shown a life of 1,000 to 1,300 hours, the progress
was rapid towards the standardization of the present series
burning lamps, and the shade, holder, and switch devices as
accessories. The description of such modern equipment and a
discussion of its operation constitutes the subject-matter of this
paper.
PAST PRACTISE.
The lighting of street cars has previously been accomplished
by using bare carbon, and in a few cars, graphitized filament
lamps. It has required from a dozen to thirty of the so-called
16 candle-power 64-watt carbon lamps in the car body, and eight
to ten similar lamps distributed on platforms and in the head-
light and designating signs. Between bulkheads the lamps
were placed about 18 inches (45.72 cm.) apart in line along the
center deck, or studded over the whole ceiling, or else grouped
HIBBEN : STREET RAILWAY ILLUMINATION 591
in clusters of four, five or as many as eight lamps arranged
radially from single fixtures on the center deck ceiling. Such
carbon lamps usually burned five in series on the nominal 550
volt power circuit, each being rated at no volts.
The high current consumption of the carbon lamps, together
with their poor illuminating performance, has led to the sub-
stitution of, first, the metallized or gem lamps, and second, the
bare 23-watt tungsten lamps in the same sockets. The former
lamps are proving unsatisfactory on account of filament break-
age from jarring and other objections,1 while the small un-
shielded tungsten lamps are but a temporary makeshift, due to
excessive glare, and because no attempt is made to utilize the
maximum amount of generated light, or direct it downward.
PRESENT METHODS.
The most modern car lighting equipment consists of one
circuit of five tungsten lamps, of the 94-watt, 78 candle-power
size, arranged in line along the car ceiling, or else an arrange-
ment of two circuits of five each, of the 56-watt, 46.7 candle-
power tungsten lamps. Quite often, in the cars where the 94-
watt lamps are used, these are placed four in the car body
between bulkheads, and one over the entrance vestibule, espec-
ially if the car is of the pay-enter type. In other types of cars,
such as the interurbans, there may be three units in the passen-
ger compartment, one in the baggage or smoking room, and one
in the vestibule. An additional circuit of five 23-watt tungsten
lamps is recommended for the large types of city cars, particularly
if these cars have the one circuit of 94-watt lamps. The
small lamps are arranged over the steps, in the headlight, and
in the illuminated designation signs.
Sometimes, but not often, the fourth size of modern lamp, a
36-watt, 26.8 candle-power tungsten-filament is used in the car
body, but the cases where the 23-watt or the 36-watt lamps are
being employed between bulkheads are largely those where no
new wiring or accessories are being installed, and where these
small lamps are replacing the carbon lamps in the old sockets
or receptacles.
1 The average life of Gein lamps in street railway service is found to be 900 hours or
less, although with the exception of a few isolated cases, the life is nearer 700 hours.
592 TRANSACTIONS I. E- S. — PART II
The four lamp sizes mentioned above are the ones that are so
far standardized for street railway service. Their characteristics
may be found in the following table :
TABLE I.— Characteristics of Tungsten-Filament Street
Railway Lamps.
Watts
Hor.
C-p.
Watts per
C-p.
Avg. Hrs
Lumeus Life
Bulb
Diam.
Overall
Length
23
17.I
1-34
168 2,000
2-78
5-#
36
26.8
1-34
263 2,ooo
2-78
5-tf
56
46.7
I.20
457 2,000
2-5/8
5-H
94
78.3
I.20
767 2,000
3-Vl6
7-H
Any of these lamps are procurable for a power line voltage
of 525 to 650, or with individual ratings of 105 to 130 volts.
They are sturdy in construction, and selected for the current
to insure a uniformity of candle-power and life.
The wiring circuits of two typical systems are given in Figs.
2 and 3. In Fig. 2 may be seen the arrangement of the 94-watt
lamps that is excellent because it insures continuous lighting
service. An extra unit is used, commonly short-circuited by a
selector switch as shown, and thus in the event of the failure of
any one lamp, (which would leave the whole car in darkness),
the manually operated selector switch may be turned by the
conductor or motorman, bridging successive lamps until the
burnt-out one is short-circuited, while meantime the extra unit
(usually in the rear vestibule) comes into circuit, and the failure
may be replaced at the first convenient point on the run.
On the whole, this wiring arrangement seems to be the most
advisable one. This scheme is all the more advantageous when
a three-way switch is wired in the circuit, and when a seventh
unit is installed in the front vestibule. A car with this arrange-
ment may always have well illuminated steps and entrances, for
in loading or unloading, the conductor may bridge a unit in the
passenger compartment, and burn both vestibule lights. Then,
whichever way the car is running, he may so manipulate the
two switches as to have a rear platform light, as well as all
lamps between bulk heads burning.
Fig. 3 shows an arrangement of two independent circuits
of 56-watt lamps, and since no selector switch is used, any one
T
5W
Fig-. 4.— One of the modern units used for street railway car lighting.
Fig. 5. — View of street car using 94-watt tungsten lamps and shades.
HIBBEN : STREET RAILWAY ILLUMINATION
593
lamp failure will leave half but not all of the car in darkness
temporarily. Fig. ia shows a portion of the rear platform with
a selector switch and one ceiling unit in place.
All lamps in the most modern street cars are being equipped
with downward reflecting shades, for reasons hereafter ex-
plained. Fig. 4 illustrates one such type of shade, together with
the holder that in reality is the complete fixture. Several forms
of holders are available ; one style is shorter than the one shown
Fig. 2.— Wiring diagram for a circuit of 94-watt or 56-watt tungsten lamps.
\
Sign
r-O-
Platform
Q
y\ Headlight
Fig. 3.— Wiring diagram for lamps in two independent circuits.
in Fig. 4. These holders all operate to clamp the neck of the
glass shade all around, with a firm grip that cannot jar loose,
and in such a way that there is no probability of breakage if
a well made shade is used.
Fig. 5 shows the interior of a typical city car, where the units
are of the 94-watt type, with a deep bowl alba glass shade. The
594
TRANSACTIONS I. E. S. — PART II
plan of this car is given in Fig. 6, from which the location and
spacing of the main lighting units can be seen.
ILLUMINATION COMPARISONS.
Two main factors are the criteria of the satisfactory qualities
of the lighting system — the cost, and the illuminating perform-
ance. The latter consideration involves the measurable amount
of foot-candles at the desirable places, and also the quality of
the light that is furnishing these foot-candle values, and its
physiological effects.
Anyone who has seen a car illuminated by the shaded lamps,
and particularly if this car has both shades and bare lamps that
can be alternately burned, will not question the fact that there
J—36 in*-«33 in^-<33 in~3 3 _in>?«33 inj-^33 in.*7-<33 iiu»f<33in.>p33".in.»f<33 ia-*<33 in-"t<33 in^io ^fr
Stations i i i 3 4 5 6 t e 9 10 II 12 '3 stations
IBf Service Lamp, 90-100 Watt tungsten
jsj Auxiliary Lamp, ordinarily not bumine
Fig. 6. — Plan of car shown in Fig. 5, showing location of lamps and test stations.
is a remarkable difference in the qualities of the light from the
two arrangements. The glare from the bare light sources is
particularly aggravating in street cars, and all-frosted bulbs
will not do much to correct the fault. Bare carbon lamps of
low brilliancy might be bearable, but bare tungsten lamps never.
The car ceilings are low, and there is a vista along which the eye
gazes. There are usually advertising cards to attract the attention
towards the upper parts of the car, and there are unavoidable
changes of the intensities from jarring and voltage fluctuations
that soon tire the accommodating muscles of the eye. Hence the
street railway lamps should unquestionably be equipped with
shades that at least protect the eyes of the passengers.
Every street railway must operate its lighting circuits and its
power circuits as one. Hence at the very time when the lights
HIBBEN : STREET RAILWAY ILLUMINATION
595
are most needed, the load on the system is the greatest, and the
fluctuations of voltage are increased correspondingly. This
trouble from voltage fluctuations was very apparent with the
use of carbon filament lamps,1 but it becomes very much less
troublesome with the tungsten lamps, since their candle-powers
do not change so rapidly at the different pressures. This can
be seen in Fig. 7. For instance, if both types of lamps gave
their full rated or 100 per cent, candle-power at 500 volts, then
when the voltage fell to such a value as 440, as it often does on
street cars, the tungsten lamps would still give 60 per cent, of their
initial candle-power, but the carbon lamps would be furnishing
Tu
ngsien
Lamps
Cart
on Lamps*
Voltage at Lamps
Fig. 7.— Voltage candle-power variation curves.
only 40 per cent, candle-power. If a lighting installation is
originally planned to furnish a certain adequate illumination at
the minimum voltage, it can easily be seen how much greater
must be the investment, and the waste, in doing this with carbon
rather than with metal filament lamps.
As to the actual foot-candle values in a typical car, note the
plotted results of photometric tests that appear in Figs. 8, 9, and
10. Fig. 8 shows the transverse and longitudinal distributions in
the car which is shown in Fig. 5. Here the typical differences
between the bare carbon, and the shaded tungsten lamps is in-
dicated. In Fig. 6 are shown the test stations at which horizon-
1 In fact over 35 per cent, of the operating companies report considerable trouble
from this source.
50
TRANSACTIONS I. E. S. — PART II
tal foot-candles were measured, at the reading height of 37
inches. In Fig. 8, the dotted lines represent the results from the
bare carbon lamps, seventeen of which were burning between
the bulkheads. These carbon lamps were in clusters, five above
test station 8-C, four above each of the three stations 11-C, 5-C,
and 2-C (see Fig. 6). Compare the longitudinal distributions
along lines A and B, obtained from these seventeen carbon lamps,
with the illumination furnished by the four 94-watt tungsten
lamps with shades, that were placed as shown in Fig. 6. The aver-
ages of lines A and B with the bare carbon lamps, are 1.9 and 2.4
-Tungstens and Shades
-Bare Carbon Lamps
m
sg
Transverse Section Station 3
Transverse Section Station 7
|
5
, lest Line B
— "
•> 4
i — — *
pN^
S
^V
. J*
\ \
o 3
est Line
x. \
>
g
\.
■ -"
~-
-^
■*,
""*"
*-
i-_ ' -"■-.'
_.
^
Line A
fl
Longitudinal Distribution . Along Test Lines A and B
Fig. 8.— Illumination curves from tests of car shown in Figs. 5 and 6.
foot-candles respectively. The corresponding averages with the
tungsten lamps and shades are 374 and 4.32. Down the aisle
(test line C) the average with carbon lamps is 2.6 and with the
shaded tungsten lamps 4.53, so that when considering the whole
car, the total averages are 2.24 for carbon, and 4.13 for tungsten
lamps with shades.1
This is the typical comparison between bare carbon lamps
and shaded tungsten lamps. The following table gives the
detailed story:
1 In averaging, double weight is given to lines A and B, to arrive at a true average for
whole car.
hibben: street railway illumination
597
TABLE II.— Illumination Data Bare Carbon vs.
Shaded Tungsten Lamps.
Bare Carbon Shaded Mazda
No. of lamps 170 4-°
Candle-power per lamp 16.3 75-°
Watts per lamp 64.0 91.0
Total watts 1088.0 364.0
Total generated lumens 357<>.o 2952.0
Area car floor 278.0 278.0
Watts per square foot 3-91 I-3I
Average foot-candles 2.24 4. 13
Useful lumens 623.0 1 150.0
Utilization factor 17-5% 39%
Relative efficiency 45 Jo 100%
As regards the illumination when small bare tungsten lamps
are substituted for bare carbon lamps, a test was made in the
car of Fig. 5, in which the arrangement of carbon lamps was in
groups as previously mentioned, and which was the same as
the arrangement of the 23-watt tungsten lamps.
*94 WATT LAMPS WITH SHADES M ON CENTER-LINE OF CAR &
X 40 WATT X ALL-FR05TED X ON SIDE DECKS X OF CAR
FEET 2 4 |6
STATIONS
£3456
Fig. 9.— Illumination curves from tests in car shown in Fig. 11.
With the carbon lamps, the averages along lines A, B, and C
were as before, namely 1.9, 2.4 and 2.6 foot-candles. The 23-watt
bare tungsten lamps gave corresponding foot-candles of 2.1, 2.6
and 2.9. The total car average illumination in the two cases
was 2.24 and 2.46, or a gain of practically 10 per cent, in foot-
candles resulting from the use of the tungsten lamps. The util-
ization efficiency of the bare tungsten lamps was 24 per cent.
A third test will show the comparison of bare all-frosted
tungsten lamps vs. clear bulb tungsten lamps with shades. Fig.
9 shows the longitudinal distribution in a car of the plan shown
4
593
TRANSACTIONS I. F,. S. — PART II
by Fig. ii, the full lines representing horizontal foot-candles
from the 94-watt lamp, and the dotted lines, the results from
the all-frosted bare 40-watt lamps. Fig. 10 shows the transverse
distribution of these latter two cases. (Fig. 10 is a section at
station 3, Fig. II.)
The car between bulkheads in this case had three 94-watt
tungsten lamps on the center deck ceiling, and eight 40-watt
round bulb tungsten lamps arranged along the side decks. These
latter unshaded lamps gave a bad glare, and were left in the
1
..M
40 W. TUNGSTEN LAMPS?
BARE. ALL-FROSTED |
i id vA
-i-7'-10/
k
25
o
ABC
Fig. 10. — Illumination of car shown in Fig. n, taken at station No. 3.
car only to afford two emergency lighting circuits. The fol-
lowing table gives the details of the tests recorded in Figs.
9 and 10:
TABLE III. — Illumination Data Bare Versus
Shaded Tungsten Lamps.
Bare Tungsten Shaded Tungsten
lamps lamps
No. of lamps 8.0 3.0
Candle-power per lamp 30.0 75.0
Watts per lamp 39.0 84.3
Total watts 312.0 253.0
Total generated lumens 2560.0 2142.0
Area car floor 1940 194.0
Watts per square foot 1.62 1.31
Average foot-candles 2.71 3.24
Useful lumens 525.0 62S.0
Utilization factor 22.08% 29.30%
Relative efficiency 75% 100%
HIBBEN : STREET RAILWAY ILLUMINATION
599
It may be interesting to note that at stations 3-A, 3-B and
3-C of Fig. 11, the foot-candles on a plane 45 ° to the rear
were 1.68, 2.34 and 2.31. At stations 6-A, 6-B and 6-C the
values were 1.97, 2.52 and 2.59 foot-candles.
A summary of the above three cases indicate that (1) bare
carbon lamps are hardly comparable with shaded tungsten lamps,
being very inefficient. (2) Replacing carbon lamps with small
bare tungsten lamps is not advisable and affords but a small
gain in the quantity of the illumination (greatly increasing the
harshness of the light), although considerably reducing the
wattage. (3) Using all-frosted tungsten calls for a con-
sumption of more energy and produces much less useful light
than shaded tungsten lamps. (4) Tungsten lamps properly
^r-9Ve-8^e'-&"w-8H-r-8W-8"-*f-«-£-8:H
* . ■ : Wv. V ■■«»■■. !w ■; Iff' ; !t? . qp — r
iT^l Jl j fTTl [p~3 [T^ fl 5 rf^l [| I £ . select)
: 4' ■' '■ + '' ': i *'■ fl ► i : *3 ;: 1
S SELECTOR
o*
-4'-0'
-5-4}-
•t* — 4-r-t
■33--e-
• 94 WATT TUNGSTEN LAMPS WITH SHADES
* 40 • • ■ ALL-FROSTED
Fig. 11. — Plan of street railway car showing location of lamps and test stations.
equipped with shades will reduce the wattage 19 per cent, as
compared with all-frosted tungsten lamps ; 7 per cent, as com-
pared with bare tungsten lamps ; 67 per cent, as compared with
bare carbon lamps ; while simultaneously in the respective cases,
the shaded tungsten lamps will cause increases in illumination
of 19.5 per cent. ; 68.0 per cent, and 84.5 per cent.
COST COMPARISONS.
. No matter how photometrically efficient a lighting system may
be in furnishing the requisite foot-candles, that system will be
inadvisable in traction service if it is not one whereby a mone-
tary saving will be accomplished, either directly by a saving of
operating costs, or indirectly through the medium of pleased
passengers and greater patronage. The new shaded tungsten
6oo
TRANSACTIONS I. E. S. — PART II
filament lamp systems can accomplish both of these desired re-
sults.
The value of a proportionately small1 saving in energy in
street railway operation is seldom given full consideration, be-
cause of its low cost of generation. But energy at the car is
more expensive than power-house costs indicate, and any sav-
ing of it through the modern system of car lighting will be
appreciable. In general, power at a car will average about 1.5
cents per kilowatt-hour. When a system of cars averaging
twenty-two 64-watt carbon lamps per car is changed over to
use five 94-watt tungsten lamps, the energy saved is 938 watts,
£
\
1
1 1
1
=?-3 S
1
1 1
;
V
1
Ave
age Daily
Peri
d of Lighting
J
v
- The time" bctv
-*i
"o "° oa
[-,
H U* '"
5000 .1000
Vtar
yAv
erag«J
Nearly Average
limi
of Lighting
Ti
11c of
Extingui
»%
4500 « 900
^
4000 "^ 800
/
k\
1
/
,
\
. 2
/
1
k
,*■
"
/~
«,
/
1
^
lot
llCa
-Hours ^
1
\
3000 -> 600
/
of iJighti
E .
j
\
/
1
,
"
\
•3
/
1
,
60
1
c
\
{'
V
I
1
/
V"
N
j /
s
/
\
V
1500 Cn 300
X
/
Ti
jl N
ot
N
/
/
1
>
Cara
in 6
mcc
1
t
j s
/
1
/
500 B 100
/
i
\
/
,
\
00 0
\
]
2
fooa
I
■
p
M.
> 1
I) 1
1 1
Hi
A.f.
L
1
0 1
1 12
Noon
Time in Hours
Fig. 12 — Characteristic curves of a typical street railway system.
amounting to 1.4 cents per car hour. If there are 1,000 cars
on the system, it has been found that there will be about
1,500,000 car hours of lighting per year, on the average, mak-
ing a total saving per year of $21,000. The cost of the shade
equipment, the excess of the cost of tungsten lamps over car-
bon lamps, and the cleaning expenses, will of course reduce this
savings perhaps 20 per cent.
Approaching the problem of economics from a different angle,
one may consider Fig. 12, which shows the characteristics of a
1 A 34-ton car with 25 passengers, consumes in its two 75 horse-power motors about 128
amperes at 500 volts.
HIBBEN : STREET RAILWAY ILLUMINATION 6oi
representative street railway system. In this figure the full-line
curve represents the total number of cars in service at any par-
ticular hour of the day. The dashed-line curve in the shape of
the hyperbola represents, for the various months of the year,
the time of lighting and of extinguishing the lamps. The dot-
ted line gives a summation of the total car-hours of lighting for
one period of darkness, being based on a yearly average. In
tabulated form, a representative lighting schedule is given as
follows :
TABLE IV.— A Typical Street Railway Lighting Schedule
Hours per Hours per
Months Period of Lighting Day Month
January- ••• 3:35 P. M. to 6:50 A. m. 15:15 472
February. . . 4:15 p. m. to 6:20 a. m. 14:05 394
March 5:00 p. m. to 5:45 a. m. 12:45 395
April 5:50 p. M. to 4:50 a. M. 11:00 330
May 6:50 p. M. to 3:45 a. M. 8:55 276
June 7:35 p. M. to 3:00 A. M. 7:25 222
July 7:40 p. M. to 3:00 a. M. 7:20 227
August 7:00 P. M. tO 3:45 A. M. 8:45 271
September . 5:50 p. m. to 4:45 a. m. 10:55 328
October • • . 4:30 p. m. to 5:40 a. m. 13:10 408
November.. 3:40 p. m. to 6:20 A. M. 14:40 440
December . . 3:15 p. m. to 6:55 a. m. 15:40 486
^Average- • 5:25 p. m. to 5:05 a. m. 11:37 354
When considering the average number of cars in service dur-
ing each hour of the dark period, and multiplying this number by
the hours each car is lighted, it is seen that, for each day, there
will be approximately 3,900 car-hours of lighting on this 1,000-
car system. This amounts to 1,500,000 car-hours per year.
Now a careful analysis made by other authorities, of power
costs of this representative system, based on the wattages of the
old carbon lamp and the new tungsten filament lamp and shade
equipment, show a total cost of lighting power per year of
$35,500 and $14,150 respectively. Therefore a summary of
lighting costs will be about as follows :
1 In southwestern United States, and in agricultural or similar regions having much
clear weather the average hours per month falls as low as 225.
602 TRANSACTIONS I. E. S. — PART II
TABLE V.— Lighting Costs, With the New and the Old Systems.
Carbon lamps Tungsten lamps
Car-hours of lighting per year 1,500,000 1,500,000
1 Average number of lamps per car- ■ • 22.38 5-94 watt 4.48-23 watt
Lamp hours of lighting per year . ... 33,600,000 7,500,000 6,730,000
Average life of lamp in hours 1,800 2,000 2,000
Yearly number of lamp renewals 18,650 3,75° 3>365
Net cost of lamps (on $2,500 contract) $2,238.00 $2,396.25 $955-65
$ 3,352.00
Yearly cost of power for lighting $35,500.00 14,150.00
2 Estimated expense of shade equip-
ment — 2,000.00
Final total yearly cost $37, 738.00 $19,502.00
The above figures, showing a probable saving of $18,000.00
or $18.00 per car per year, compare closely with the estimate as
previously presented, and moreover they represent a very con-
clusive argument for the adoption of the modern lighting sys-
tem. This is but the financial viewpoint, and one should not
loose sight of the accompanying increase in the amount of light
(84 per cent, increase, see table 11) nor of the better qualities
of the illumination.
SUMMARY OF THE STATUS AND PROSPECTS OF
CAR LIGHTING.
Briefly, the most important considerations to the gathered from
the above discussions are summarized as follows:
The bare carbon or graphitized filament lamps are no longer
advisable nor economical for street railway service, on account
of short life, high energy consumption, poor quality and unsteadi-
ness of light, and the wiring costs incidental to the installing of
a large quantity of them.
Small bare tungsten lamps replacing them are nearly as waste-
ful from the standpoint of not having their light directed by
proper shades, and in addition are extremely dazzling and
optically harmful.
Large tungsten lamps, of 94 or 56-watt sizes, seem the most
1 The fractions of lamps come from an average of motor-cars and of trailers, the latter
having fewer lamps, and in the case of the tungsten lamps, having but the one circuit of
five 94-watt units.
2Basedon (1) First cost $7, coo, or interest of $2So.oo. (2) 10 per cent, breakage and
depreciation, or $700.00. (3) Cleaning twice monthly $1.05 per year per car or $1,050.00. As
regards this cleaning expense, the data is meager. A fair figure for dusting once daily
seems to be 60^ per car per month, and washing twice monthly, 6o# to $1.25 per car per
month.
HIBBEN : STREET RAILWAY ILLUMINATION 603
economical, in conjunction with shades. These units may more
than double the previous (most generally inadequate) values of
illumination, at one-third of the energy cost, giving a soft dif-
fused light and forming a neat decorative unit.
The candle-power of the tungsten lamps is much more steady
and less affected by voltage fluctuations.
The net cost of lighting with the new units is much less than
when carbon lamps are used; the cost of the tungsten lamps is
slightly greater, but this is more than compensated for by the
energy saved.
All these items, and others that appear from a study of this
subject, seem to indicate a prospect — almost a certainty — of a
rapid and a large development in street railway car lighting.
Up to the present time there have been perhaps a half dozen
installations of large tungsten units, and shades. Bare tungsten
(together with a few tantalum) lamps are in service on approxi-
mately 28 per cent., bare graphatized lamps on 10 per cent., and
bare carbon lamps on 60 per cent, of the 175,000 to 200,000 cars
in service in the United States.
Very soon the traveling public may be expected to demand
better lighted cars. As the passenger looks lengthwise of the car
(75 per cent have cross seats) or up at the advertising cards (on
95 per cent, of all cars) he cannot help but be affected by the
glare from the old lighting systems. If he may be freed from
this annoyance, and does not have to strain his eyes because
of insufficient or fluctuating illumination, surely he becomes a
more valuable asset to any railway company.
Therefore this modern lighting system seems destined to be
mutually advantageous to street railway companies and passen-
gers. It is hoped that further investigations of this subject
will soon afford valuable additions to the somewhat preliminary
study that this paper presents.
604 TRANSACTIONS I. E. S. — PART II
DISCUSSION.
Mr. L. C. Porter: A little over two years ago Air. Stickney
took up the question of street car lighting. We went to several
railway companies to put the proposition up to them. The first
question that arose was : Will the lamps stand the severe use
of street car service? At that time we had results of tests which
had been conducted on some of the United States battleships
where tungsten lamps had gone through target practise, full
power run and several other severe tests, giving very good re-
sults. We also had tests on ferry boats where the lamps oper-
ated satisfactorily.
With the co-operation of Mr. Hoist of the Bay State Street
Railway Company, we equipped some of their cars with 56-watt
tungsten lamps and prismatic reflectors. The question of the
different sized lamps was studied and it seemed best to use
the 56-watt lamps for several reasons. The principal reason
was that with the 94-watt lamp the failure of one lamp would
put the car in darkness for at least a few minutes until the con-
ductor could find the selector switch and operate it; while with
the 56-watt lamp there were two circuits, and since the lamps
are wired on alternate circuits the failure of one lamp would
not plunge the car into darkness. Another advantage which was
found in using the 56-watt lamps was that the lamps could be.
spaced a little closer together, giving less sharp shadows than
are likely to maintain in the 94-watt system. For instance, look
at the diagram on page 598; and consider a passenger seated in
the second seat from the left-hand end of the car, holding his
paper on what is an average reading plane; that is, 450 (3 feet
above the floor). When facing the right-end of the car the
passenger will get very good light on the paper, but when he faces
the left-end the illumination on the paper will not be so good,
because the light which is then shining on it will come from a
considerable distance; whereas if a lamp were placed halfway
between, there would be practically the same amount of illumin-
ation no matter whether the reader were facing forward or
backward.
Twenty-eight-foot cars and 34-foot cars were equipped with
56-watt lamps. In installing this system we spaced the lamps 6
STREET RAILWAY ILLUMINATION 605
feet apart, in this way bringing the two end lamps 2 feet from
the ends of the car, with another lamp on each platform, one
in the headlight and one 56-watt lamp in each sign of the car.
Where two lamps had been used per sign, we installed 2 28-
watt, half voltage lamps (equivalent to one 56-watt, full voltage
lamp), thus making two full circuits of 5 56-watt lamps each.
Photometer tests were conducted in cars so equipped and it
was found that in a car that had previously been equipped with
25-watt 64 carbon lamps, consuming 1,600 watts, we reduced
this consumption to 560 watts by using the 56-watt lamps which
gave approximately 3 effective lumens per watt; whereas only
0.4 was obtained with the carbon lamps.
Both intensive and extensive type reflectors were tried. It was
found that the intensive type of reflector gave the better results.
It gave a better average intensity and threw the maximum
amount of light where it was most needed, that is, over the
seats.
An interesting shadow test was conducted in which Mr. Hoist
seated himself in various seats in the car and was surrounded
by a number of men, to see if they could cast objectionable
shadows over the paper he was reading. It was found impos-
sible to do this.
Six of these cars were operated by the Bay State Company over
various conditions of roadbed, continuously for over a year in
order to make absolutely sure that both the reflectors and the
lamps would stand the service. At the end of the year they had
proved very satisfactory. Those tests were reported by Mr.
Hoist in the Electric Railway Journal for September, 1912.
As a result of the tests the Bay State Street Railway Company
is equipping all of its new cars with this system of lighting and
also rewiring some of the old ones. There are several other
companies throughout the country that are doing likewise, prov-
ing that this system is a paying proposition.
Mr. E. B. Rowl : Mr. Hibben's paper is an excellent one
for our Transactions because it adds something new. Street
car lighting, or rather the proper illumination of street cars, is a
relatively new development and there are tremendous possibilities
in the way of improvements.
606 TRANSACTIONS I. E- S. — PART II
During Mr. Porter's discussion it occurred to me that circum-
stances may alter cases : this should be remembered in discus-
sions of the 56-watt 2-circuit system versus 94- watt 1 -circuit
system, from the standpoint of lamp failures. The consequences
of the total though brief loss of light in a car which would result
when a lamp fails in the latter system would depend to a large
extent on whether it is totally dark outside of the car and con-
sequently on the territory over which the car operates. In most
city streets there is enough external light coming in so that the
total failure of the light in the car is not of so much importance,
whereas in running over a private right of way or over country
roads it is of great importance.
Again, as regards ability to read with the one-circuit and two-
circuit systems, — that may depend on the type of car construc-
tion. If it is a car having the passengers all facing in one direc-
tion and not across the car, the need for good reading illumina-
tion, both in intensity and proper direction, is greater than in that
type of car which is rapidly coming into favor for con jested city
service. My point is that what has proved to be the best type
of lighting system for one installation is not necessarily the best
type to adopt for some other traction system.
In Mr. Hibben's paper he mentions in his introduction that the
first installation of lamps with individual reflectors in street cars
was in 1909 in Dayton. I believe there were several isolated
instances where single carbon lamps were equipped with pris-
matic reflectors prior to that date. There is one such installation
on the Washington & Baltimore Traction System.
During the past few years or even months, to one who has
been following the situation, it is quite remarkable to note the
change in attitude regarding this question of car lighting. Sev-
eral years ago the question of changing the lighting in the subway
cars of New York came up under the direction of the Public
Service Commission. Extensive tests were made to show what
improvement could be effected by change in equipment. These
tests brought out clearly of course the advantages from the stand-
point of uniformity and higher intensity of using reflectors over
the lamps and by substituting tungsten lamps for the carbon
STREET RAILWAY ILLUMINATION 607
lamps then in use. The danger from glare, however, was not so
well understood at that time and the use of reflectors was tem-
porarily decided against because, as one of the managers of the
subway company expressed it, "the public was not educated to
the need of reflectors." This is an attitude which will probably
always require a great deal of effort to overcome. Something
really desirable should not be turned down because the public
has not been "educated up to it." I believe the change should be
made and the advantages proved in actual practise.
The use of a spare lamp with the selector switch gives a very
satisfactory system of illumination. On the Cleveland cars the
newer ones have the extra lamp placed in a similar fitting to the
lamp in service over the rear platform, these two lamps being
placed close together in individual porcelain enameled bowl-shaped
reflectors which are counter sunk in the ceiling of the car. These
reflectors with the prismatic reflectors used in the body of the
car provide a very satisfactory system giving practically no glare
effect; and the steps, platform and seats are amply lighted.
As these are pay-enter cars the selector switch is always within
easy reach of the conductor and only a few seconds are required
to locate the dead lamp, in case of a failure, and replace it by
the spare lamp at the first convenient opportunity.
The effect of changing from carbon to tungsten lamps from
the standpoint of change in intensity with voltage variations is,
I believe, quite noticeable to anyone who has used a line fre-
quently before and after such a change. On the Columbus-
Zanesville Interurban Line I had an opportunity of observing
this fact and while I knew what to expect I was actually as-
tounded by the improvement after the substitution of the tung-
sten lamps. Before the change when a car pulled out of the Co-
lumbus station it was necessary to stop reading, roll up one's
coat for a pillow and try to get some sleep ; the voltage dropped
so much that the car was almost in darkness. After the tungsten
lamps were put in one could read fairly comfortably during the
entire trip, even though the voltage fluctuation on that line is
quite extreme.
The use of auxiliary emergency lighting might be mentioned
because this ought sometimes to be provided on lines where an
608 TRANSACTIONS I. E. S. — PART 11
absolute or extended failure of the service might result in panics
and serious accidents. The use of a small storage battery equip-
ment with a few lamps placed at convenient points in the car
will provide light instantly when other sources fail.
It may be interesting to the I. E. S. members to follow a series
of tests in a study of coach lighting which has just been conducted
under the auspices of the Railway Electrical Engineers. Al-
though these tests were conducted with primary reference to
steam road conditions, the illumination requirements in coaches
are so similar to those in street and interurban electric cars that
these tests should be of considerable help in designing the light-
ing systems for electric cars. These tests were quite extensive,
involving a consideration of shadow effects, etc., and will un-
doubtedly be published in full shortly.
Mr. G. H. Stickney : I believe thoroughly that the use of re-
flectors with tungsten filament lamps in street car lighting is the
coming practise, both on account of the superior lighting effect
and economy. (The installation of such units on the Bay State
Street Railways, described in the Railway Electrical Engineer
of September 28th 191 2, was the first of the recent installations
of this type.) On the other hand, I cannot quite agree with the
author that the use of the bate tungsten filament lamp is not justi-
fied, at least as a temporary expedient, for certain conditions. In
the first place, on account of the expense and time required to re-
wire and equip old cars already in active service, it is impracticable
for many roads to change over completely immediately. In the
second place, the demand for soft, diffused light in street cars
is nowhere near as great as in train lighting or other interior
lighting installations, since the passengers are not in the cars for
relatively long periods of time, and, wearing hats, they do not
have their eyes exposed to the glare to the same extent as under
these other conditions. I have ridden in the subways in New
York, which, as you know, are equipped with bare tungsten fila-
ment lamps, and it has been my observation that relatively few
of the passengers have been conscious of the glare effect, while
many have appreciated the higher intensity provided and the
improved steadiness of the light under voltage variation, jwhicli
STREET RAILWAY ILLUMINATION OCX)
is a very decided advantage introduced by the tungsten filament
lamps, whether with or without reflectors.
I firmly believe that the question of economy alone should
induce the railways to immediately adopt reflectors in all new
cars, and that in the long run it will compel their adoption even
at the expense of re-wiring, for all roads in which costs are
carefully figured.
Mr. Porter has mentioned tests in which we are assisting the
New York Municipal Railways. I hope that it will be possible
to make these tests public, as they are probably some of the most
complete yet undertaken. Practically every arrangement of light-
ing that promises to be suitable for the subway cars in question
is being tried out and observed.
Referring again to the paper, the author states that there are
half a dozen lines using tungsten filament lamps. I think this
is a little conservative, as I am sure we have handled a consider-
ably larger number than that in our own office.
Mr. V. R. Lansingh : The substitution of tungsten lamps
for carbon lamps naturally increases eyestrain, and it is a
question with me whether or not the economy gained by the use
of the new lamps and the increase in illumination is not more
than offset by the decrease in eye comfort. Furthermore, when
the railroad has made such substitution the economy to be gained
by the use of reflectors is smaller than it would have been if
they had started in at the beginning and changed the entire sys-
tem. It is a question, when the new lamps have once been in-
stalled whether there would ever be a change to the old wiring
system and in new lamps and reflectors installed. I believe I
would rather see the old system remain in place and the new
complete system will therefore come sooner.
Mr. C. W. Bettcher: I believe that the fluctuation in the
candle-power of carbon lamps, due to the change in voltage, is
much more objectionable than the glare from the tungsten fila-
ment lamps when they are installed. Most people do not care so
much about the glare; in fact they do not notice it when reading;
but certainly the change in candle-power is very objectionable,
especially where a good many stops are made and frequent start-
ing causes a temporary drop in voltage.
6lO TRANSACTIONS I. E. S. PART II
Mr. Ward Harrison : In comparing the relative advantages
of the 56 and 94-watt lamps, the type of car should be considered.
Since in Cleveland pay-enter cars are used almost exclusively,
the short circuiting switch is always within easy reach of the
conductor stationed at the fare box and a burn-out in the car
will not cause the lamps to be extinguished for more than a few
seconds. If on the other hand the type of car is such that the
conductor is not always at the same place, there is surely an
advantage in having two circuits.
Mr. R. B. Ely: Within the past two or three years a num-
ber of papers have been presented on car lighting. One of these
has advocated a system by which the light would be directed
downward and forward, the light source being semi-concealed.
Such a system would be an improvement over the present method
of lighting the near-side car. New cars of this type are fully
equipped with illuminated signs, illuminated steps, etc. The
lighting installation consists of bare lamps placed along both
sides over the seats. A semi-concealed system of lighting could
possibly be used to advantage in this type of car.
Mr. J. B. Jackson : There is one item that I believe should
be brought out more clearly in a cost analysis and that is first
cost of equipment. This is a point which is of great importance
to the street railway company and one which will be the prin-
cipal consideration in changing the lighting equipment in present
cars. It seems as though Mr. Hibben's cost of $7.00 per car is
rather low as that would indicate a cost of $1.17 per unit for the
six-unit equipment. I believe with the special holders required,
the simplest equipment possible, i. e. holder, socket and reflector,
could not be installed for much less than $2.50 per unit. This
will make a slight increase in the last item of the cost analysis
Table V. I would also suggest that the words "per year" be
added to the item making it read "estimated expense of shade
equipment per year."
Mr. S. G. HibbKn (In reply) : Concerning shadows, re-
ferred to by Mr. Porter, I have in mind one case where the ob-
jection to lighting with the center-deck units was put forth with
the effect that a standing passenger would shield a seated pas-
STREET RAILWAY ILLUMINATION 6ll
senger, particularly if the passenger were seated on a longitudinal
seat. In a brief investigation I found that there are ordinarily
no grounds for such an objection. The light from the units
at one side or the other of the standing passenger will give suffi-
cient illumination so that there will be no sharp shadows on the
seated passengers.
Formerly the proposition has been advanced that center deck
units would not light the advertising cards. In actual practise
the center of the light sources, if this type of unit were used,
(referring to the fixture that projects entirely below the deck)
would be 6 or 7 inches below the car ceiling, and there would
be full illumination of all sides. In fact in all cases that
I have seen, one is better able to read these cards, and with more
comfort. One can hardly see at all the sign that is directly be-
hind a bare lamp on the edge of the half deck.
Mr. Harrison brought out the point that on certain cars per-
haps the conductor would not be within reach of the selector
switch. In that case one solution might be to place the switch
in the motorman's end of the car (if it were not of the double-
end type) so the motorman would always be able to quite con-
veniently turn it.
Mr. Ely brought out the fact, and a rather surprising fact it
is, that a large number of cars have improvements in heating
and ventilation and illuminated signs and headlights, while as
yet not much attention has been given to improving the illumina-
tion between bulkheads.
I will not enter into any lengthy discussion with some of
these lamp men, for I believe they will themselves wish to re-
consider or qualify the remarks as to the use of unshaded lamps.
I advocate reflectors because in the first place I believe that
through their use they will pay for themselves. It is purely a
matter of economics, because if a reflector can increase the use-
ful light, as it certainly can in most, cases, it follows that the
wattage may be reduced, and that brings about a saving that
will more than pay for the added expense of the reflectors and
their maintenance.
Regarding Mr. Jackson's statement concerning the costs of
units, — possibly these figures of the first cost of $7,000 — about
6l2 TRANSACTIONS I. E. S. — PART II
$7.00 per car — are a little bit low at the present prices of these
units. The price of these will drop somewhat, as their use is
extended. The cost will run about $1.17 per unit for the fixture
and shade. I do not consider the installation cost a factor, since
the wiring will be about as expensive for the large number of
the old units, as for the less number of new units.
ELY: CHURCH LIGHTING 613
CHURCH LIGHTING.*
BY ROBERT B. ELY.
Synopsis: This paper discusses some of the problems encountered in the
lighting of churches. It outlines briefly some requirements of certain parts
of churches of different denominations. Views of typical interiors are given.
I. STRUCTURE.
In discussing the illumination of churches it may be apropos
to give a few facts concerning the history of the early ecclesias-
tical structures as places of Christian worship.
The structures were not copied from either the Heathen or
Jewish temples, but from a combination hall of justice and a
market place, which was called a basilica by the ancients. The
rites of heathendom were performed almost exclusively by the
priest, and the temples were lighted only by the daylight that
came through the doorways or interior courts, or by a few lamps
that burned around the image of the God. The temple was not
regarded as an assembly room for worshippers, but only as an
abode of the God. Thus the dark and mysterious temples were
unsuited for religious services, in which the people were to
participate. Although the basilica served its purpose as a
place of worship there was little or no significance in the
structure to awaken Christian sentiment. The Christians from
an early period used the cross as a sacred emblem, and in their
endeavor to indicate their allegiance to the author of their salva-
tion they modified their structures to the form of the cross ; both
the Latin and Greek crosses were followed. In either case the
arms at right angles and directly opposite each other, cut it
across, and were given the name of transepts. Over the point
of intersection of the transepts, the body of the cross, a central
tower or spire was frequently erected. Beyond the galilee or
entrance chapel, or in some instances the entrance door to the
transepts, was the nave. If there were no transepts the nave
would extend from the choir to the principal entrances, but
would not include the aisles. Side aisles frequently continued
along the transepts, thus running around the whole church ;
sometimes there were double aisles to the nave. Beyond the
* A paper read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
5
614 TRANSACTIONS I. E. S. — PART II
transepts was the chapel or chancel, in which was situated the
altar; sometimes there were several altars. Side chapels will
sometimes be found on the side aisles.
The early Christian churches were lighted by daylight through
the construction of a clear story.
MODERN CHURCHES.
There are scarcely two churches alike in structure although
there is often a similarity of the plans of churches of the same
denomination. For example, the Roman Catholic Church has
generally adopted the Italian Renaissance and the Episcopal the
English Gothic. But there are comparatively few classic ex-
amples in existence ; most of the churches of the present day are
modern adaptations of several styles of architecture. The material
used in building construction and the building law requirements
have made it necessary to depart from the purely classic styles
in order to provide for fire-proofing and other construction
details ; so that one no longer sees a style or order in its true pro-
portions. The endeavors of an architect, therefore, are not
strictly along the lines of what has gone before. He usually
attempts to work the adaptation of one of the various styles of
architecture into a pleasing ensemble.
Any building set apart for religious services is termed a church,
excepting those buildings of smaller dimensions, which are called
chapels.
The Gothic style of architecture predominates in the construc-
tion of present day churches. This style is considered one of
the noblest and most complete in architectural design. Its
distinctive features are the Gothic pointed arch, the tendency
toward vertical lines, deep mouldings on columns, capitals, etc.,
and decorations derived from various kinds of foliage. The
towers are frequently square at the base and terminate with
lofty spires richly decorated. The hammer beams and pendants
are also among the chief features. But in general the modern
churches are adaptations of the French and Italian Renaissance.
Intensity of Light. — In considering the quantitative values of
the illumination in the church, experience has shown that it is
inadvisable to stipulate a certain intensity of light for the audi-
torium, the sanctuary or chancel. It is advisable, however, to
£>H
Fig. i. — Direct illumination of a church.
Fig. 2.— Overhead illumination in a synagogue.
Fig. 3. — Cove lighting in a church.
Fig. 4.— Indirect lighting in a church.
Ely: church lighting 615
determine a relative ratio between the intensity of light of the
chancel, and that of the main part of the church. That ratio in a
number of effectively illuminated churches is about 2 to 1 or
greater in favor of higher illumination in the chancel or sanctuary.
Owing to the general systems of control, particularly of electric
lighting installations and the manipulation of this control, prior
to the beginning of services one-half of the installation is usually
in use. During the general service, in which the congregation
takes part, the entire chancel and auditorium is illuminated.
While the sermon is being delivered the illumination in the audi-
torium is usually reduced, and the chancel or such lamps that are
used to illuminate the pastor and pulpit are used. The entire
equipment is again used during the closing services. Provision
should be made in Roman Catholic churches for illumination of
the stations of the cross throughout the day and evening.
Any set calculation relative to watts per square foot is inad-
visable, owing to the numerous variables found in church struc-
ture. Some churches are illuminated with an energy consump-
tion of 0.3 of a watt per square foot of floor area, while others
require as high as 2.5 watts per square foot, both installations
being considered good examples.
There is, no doubt, greater intensity of illumination in the
newer and reconstructed installations of churches ; yet the varia-
tion in intensity of illumination is comparatively wide.
Commercial factors such as costs of installation and operation
present themselves in all but a few instances, and tend to deter-
mine the character and intensity of the installation to a great
extent. A very elaborate lighting equipment can be designed,
but unless the bearing of the commercial factors of the case has
been determined, the chances are that the plan will be discarded.
In the illuminating engineering work of the Philadelphia
Electric Company a very broad policy permits the lighting
specialist to draw up plans and specifications that are guided
largely in each case by the church . officials, or architect. The
company aims to present a proposition that will be in keep-
ing with the architecture, effective in results from an illumi-
nation stand-point, and economical in operation. The specialist
treats the proposition in an unbiased manner. He is permitted
to specify any system, shades, reflectors or reflecting devices,
6l6 TRANSACTIONS I. % S. — PART II
which in his judgment will meet the conditions of a given case.
Due attention is also paid to possible emergency lighting by gas
units.
It may be of interest to mention some of the demands and
tendencies of the clergymen and architects in such cases. In
Philadelphia many architects have presented their plans to the
company with instructions to lay out an indirect lighting propo-
sition. In numerous instances where a direct lighting system
has been laid out by the architect, the church authorities have
brought their plans to the company to have estimates fur-
nished for indirect lighting, in spite of the architect's drawn
plans. There has been practically no call for semi-indirect light-
ing, but plans for semi-indirect lighting have been drawn and
recommended where, in the opinion of the lighting specialist, the
conditions were more favorable for this method of lighting.
Lighting Systems. — Direct lighting having been used to the
greatest extent still predominates. This is largely due to the
architects, who seem to be more familiar with this method of
lighting. (Fig. i.)
There is a tendency, however, in favor of indirect lighting, or
a concealed direct lighting system.
The semi-indirect system has not been adopted to any extent,
owing to the greater cost of translucent glass bowls of large
dimensions, and the greater installation costs.
Direct lighting systems have been installed more generally, due
to lower installation and operating costs, the influence of the
architect, and the lack of information pertaining to lighting
matters on the part of the general public.
I do not mean to favor any particular method of lighting;
the foregoing statements are based on installations that have been
made recently.
It is not at all unusual to see new buildings and churches
equipped in a manner at variance with good practice, simply be-
cause the architects plans were drawn and the client had entire
confidence in the architect. Generally architects will co-operate
with the illuminating engineer but this is not always the case.
r
.
\&yz~- 8*11 1 1
n -f IF iln \m l<Sm ;-
9 l^2iHfiMm lis y»^-Ja
i *H ■ ill *i I I*7" 1 I II
i Irai ]|
11 IIh.1I j»SH*!if!!
wk:
Bf '■■
$^&mb
GH^'^vTTI
MM j 1
H| 1
[■Wcaraa} v* «o40
flr ^
-W^^BM— 1— ta—r l \ j$k
u^B BP W-fll 1 a
|7 *j
ft ^
Mfjrl;
ill— -__ ' " llA^'A^^'4'^"tff ■
Fig. 5. — Altar illuminated with tungsten -filament lamps.
Fig. 6. — An electric church sign.
5u
Jo/ June 8f
(LDRCNS 0/J
it
19
m^k
ii._t.-.y_-
1
Take P.iit.
K>:-Z-7?
COOP
5/M7/.VC.
'. \Su 1
Fig. 7.— Sign over church entrance.
Fig. S. — An illuminated bulletin board.
Fig. 9. — An Illuminated box sign.
Fig. 10. — Lighted entrance to church.
Ely: church lighting 617
General Requirements (Auditorium). — There are some gen-
eral rules that should be followed, relative to the illumination of
the church auditorium.
No light sources should be within the line of vision.
All lamps in any system of illumination should be hung suf-
ficiently high, so they will not obstruct the view of those in
the gallery of the church.
No fixtures should be so located that they will detract from
the architectural features.
The intrinsic brilliancy of all light sources visible from any
point should be reduced, by frosting or other means.
Hymn boards should be illuminated by a concealed source.
Chancel. — Generally in the construction of the chancel one or
more windows or sky lights are provided for daylight illumina-
tion of the chancel, thereby making this section of the church
lighter by comparison than the main body of the church. With
artificial illumination this effect may often be obtained more
readily and more effectively.
In formulating a plan or method of lighting the chancel, or
sanctuary, the construction of the interior of the chancel will fre-
quently demand special apparatus, reflectors and lamps, in order
to obtain effective results. A chancel arch will greatly facilitate
the installation of lamps and reflectors, but when there is no arch,
it is more difficult to place the lamps and reflectors to direct the
light to the best advantage and avoid sharp shadows. Such
equipment when placed at a great height should be so arranged
that it can be lowered for renewal and cleaning purposes. But
owing to the installation costs lighting units are more often
located in a permanent manner behind the arch and renewals are
made by the use of extension ladders.
Altars. — The use of decorative altar illumination is increasing.
Electric imitation candles are being used in many churches.
The lighting of tabernacle niches as well as the outlining of
altars is being done quite successfully by the use of low voltage
lamps.
A sign lighting transformer is often installed and receptacles
6l8 TRANSACTIONS I. E. S. — PART II
provided at various points on the altar, together with the perma-
nent equipment, so that small electric bulb signs and emblems
may be readily connected. The cost of such an installation is more
or less offset by the low operating costs of the equipment. For
instance in the candelabrum in Fig. 5 there were 100 8-candle-
power 34-watt carbon lamps operating at a cost of about 34c.
an hour. They were not used, except for special services, on
account of the cost of operation. Upon installing a sign lighting
low voltage transformer and 5-watt tungsten lamps the cost was
reduced to about 5c. per hour. This resulted in greater use of
the equipment. Of course the illumination was reduced, but this
was offset by the installation of two trough reflectors with
tungsten lamps behind the sanctuary columns.
On account of the tendency in recent years to increase the il-
lumination in the body of the church it is necessary to increase
the illumination in the sanctuary to make a noticeable difference
in the intensity of the light in the two sections of the church.
The depth of some of the chancels is fairly great and in order
to increase the illumination on the altar and reredos it is necessary
to concentrate the light from a source more or less localized.
Apparatus and Fixtures, etc. — The use of dimmers in the
church is becoming more general. In some Episcopal churches
they are placed on a circuit of imitation candles. As the choir
procession enters the chancel, the lamps are gradually brought to
full brilliancy. In some of the Catholic churches dimmers are
connected to the circuits of the church auditorium and sanctuary
for use during special services.
The question of appropriate fixtures is an important one.
The illuminating engineer should have a knowledge of decoration
and ornamentation, for he is often called upon to not only plan
the outlets for a lighting scheme, but to select the fixtures. In
the case of some of the more novel schemes of illumination, he
is obliged to design fixtures to meet unusual conditions. This
means that he must be familiar with the different orders of
architectural design. The illuminating engineer should study
harmony of design, and particularly the effect of light and shadow
on architectural ornaments.
There have been many apparent reasons given for copying
Ely: church lighting 619
ancient lamps and light sources to carry out certain styles of
architecture. In carrying out such designs of fixture and lighting
units, the units have frequently been so exposed and located
in such a manner as to be distressing if not harmful to the eyes.
Therefore, novel and unusual methods may be inaugurated in
designing the illumination for the church with due regard for
ritualistic symbolism.
When the illuminating engineer attempts to effect new and
unusual methods of lighting churches he should of course be
familiar with the orders of architecture, and ornament, so that
he will be able to so design the lighting installation that some
of the architectural features of the building may be brought out
by light and shadow without distorting their appearance.
The candle is used in the services of the Roman Catholic,
Episcopal (high), Lutheran and German Lutheran churches.
It is given more prominence in the ceremonies of the Roman
Catholic church than in the others.
In the Catholic churches the altar is the main feature to be
considered. A rubric of the church requires the use of wax
candles for various services. However, in numerous instances
some of the candles that are still in use can be replaced by either
gas or electric imitation candles to greater advantage, and this
will result in a more effective appearance.
The lighting specialist should know when and how candles
are used in religious services, their advantages and disadvantages.
The mellow glow of the candle with the low intrinsic brilliancy
of its flame is pleasing, aside from the flicker. Whether the
flickering of the candles for religious purposes is a distracting
feature, or whether the candles are more effective under such
condition is probably an open question. There are certain times,
particularly on holy and feast days, when the altar is illuminated
to a greater extent than for ordinary services.
The candle has some disadvantages that cannot be overlooked,
such as flicker, fire hazard, and labor required to maintain the
proper appearance of the candle and fixture. Therefore, all
candles that are used in addition to those required by the rubrics
of the church may be replaced with electric imitation candles, or
gas burners.
620 TRANSACTIONS I. E. S. — PART II
Organs. — The lighting of the keyboard and pedals and interior
of the organ frequently present some difficulty, owing to the
location of the keyboard and the necessity for concealing the
light source from the eyes of the congregation. Some lighting
installations have been marred by the glare of an exposed light
source supplying the illumination for the music rack and key-
board, and by the lamps that are used to illuminate the choir
when it is located in the front of the church.
It is customary to provide a portable lamp or extension bracket
for the music rack connected to a live circuit to permit the organ-
ist to illuminate his music during his rehearsal on weekdays
without using a number of lamps.
Illumination for the choir, when it is located in the gallery,
can be provided by brackets or ornamental standards in direct
lighting installations.
A lamp equipped with a metal reflector located under the key-
board is frequently recommended for the illumination of the
pedals.
ADVERTISING AND EXTERIOR ILLUMINATION.
Church advertising is being indulged in. A number of churches
in Philadelphia purchase half a page of the newspapers on Sat-
urday of each week, and announce subjects, relative to church
benefits and aims, that are to the interest of the community.
This advanced step towards business management of the church
has greatly assisted in the introduction of exterior illumination of
the church and the adoption of the electric sign. Numerous
churches have installed both illuminated and straight electric
signs. Some of these signs contain large numbers of lamps.
Greater attention should be paid by the manufacturer of elec-
tric signs to their designs. The sign should be of a design some-
what similar to the architecture, and may be improved by the use
of Gothic or old English letters. The use of these letters in
themselves would give to the sign a certain dignity that is not
obtained with the block letters that are in general use. But the
legibility of a sign might thus be somewhat sacrificed. However,
this is a detail which should receive more favorable consideration
from those interested in the introduction of signs.
The illuminated cross on the steeple of the church is' very
ELY : CHURCH LIGHTING
621
effective, and can be seen at a great distance. There is no doubt
that it is an advertiser of the church.
Mr. A. Larney describes in an article on the erection of an
illuminated cross, a very interesting method of installing the
lamps :
Lamps are screwed into the receptacles mounted on three
flexihle belts equipped with a series of rope guides and pul-
leys, by means of which the installation can he lowered from
the cross down into the interior of the spire. In this way
lamp renewals can he made without the assistance of a
steeple-jack or building a scaffold.
Fig. 11.— Church exterior showing lighting standards.
In spite of the many drawbacks and difficulties involved in
erection and maintenance of such crosses, a great many of them
have been installed. The renewal of lamps lias, in numerous
instances, been the work of a steeple-jack, but the construction,
as described by Mr. Larney, may be possible in numerous in-
622 TRANSACTIONS I. E. S. — PART II
stances, and should lead to greater use of this method of adver-
tising the church.
The use of lamp posts to improve the illumination of side-
walks about the exterior of the building tends to make the
churches stand out to better advantage, and to encourage attend-
ance, and adds to the safety of the pedestrians when the side-
walks are covered with snow or ice.
Bulletin Boards. — Some churches have bulletin boards that
may be illuminated at night. Special features of this kind add
to the value of church lighting. A bulletin board of this kind
can be illuminated every night at very small expense to the
church.
The illumination of clock-dials is a feature that should not be
overlooked, for it adds to the nightly use of light by the church,
and acts as a constant reminder to those traveling in the vicinity.
Entrance to Church. — Entrance lighting is important. It not
only illuminates the stairway and the approach to the church,
but indicates to the general public when illuminated that services
are being held. Ornamental brackets containing lamps are used
extensively for such lighting. Often the arch of the entrance is
outlined with a series of lamps.
The importance of proper church lighting is too often min-
imized by pastors, church boards, and those who have to do with
church management. Ordinary business methods are about as
important in the management of the church as they are in a
mercantile establishment ; and perhaps more important because
the objects to be obtained in the case of the church are so much
more vital to the entire community. It is the business of those
who manage church property to see to it that attendance at
church services shall reach a maximum, and remain there ; and
yet it is surprising how little is done to make our churches
inviting, cheerful and comfortable. There is no single factor,
which will accomplish this result so surely, and at such small
initial expense as a scientifically designed system of lighting.
Every one knows that any illumination which produces eye-
strain or fatigue among the congregation is a positive force tend-
ing to reduce the interest of those attending church services. Pro-
prietors of places of amusement discovered this fact many years
ELY: CHURCH LIGHTING 623
ago, and now-a-days they look upon the use of light as a posi-
tive necessity.
From every standpoint good illumination is just as important
in the church as it is in the place of amusement. Emphasis is
laid upon the comparative methods employed by places of amuse-
ment and churches in the matter of making their audiences com-
fortable. There is nothing in religion which teaches that
those who attend church services should be made uncomfortable,
and yet lack of forethought upon the part of those responsible
too often makes attendance at church services a positive source
of distress for those whose eyes are not strong or cannot stand
the eye-strain caused by poor illumination.
DISCUSSION.
Mr. J. R. Cravat h : I think it is very gratifying that one
member here in one year is able to present so many comparatively
excellent examples of church illumination. I wish to call your
attention to some of the future possibilities of cove lighting with
the new helical filament tubular lamps. The cove lighting shown
by Mr. Ely shows some undesirable characteristics of cove light-
ing as it has necessarily been carried out in the past; that is an
excessive amount of light close around the curve with too little
farther away. The new tubular lamp with the helical filament
is going to make it possible to control the light much better for
cove lighting and without some of those bad effects that have
heretofore been inherent in it.
Fig. 1, shows a church with paintings on the ceiling. I am
not prepared to say that there is any better way of lighting that
particular church, but I simply want to call your attention to
the fact that any exposed lights are going to hide the paintings
to a certain extent, and that this fact must be borne in mind by
the architect when he is designing the church. If it is so de-
signed that it must necessarily be lighted by some exposed light-
ing unit, the effect of the paintings will be lost at night.
Mr. A. L. Powell: Church lighting forms an extremely
interesting subject and an observation of most of the existing
installations shows that there is still a great deal of work to be
done.
624 TRANSACTIONS I. E. S. — PART II
By way of supplementing the data included in the paper, the
following remarks may be of value.
As a result of a number of tests, it has been found that an
average intensity of 1.5 foot-candles on a 3-foot horizontal plane,
is very satisfactory for church lighting, where proper precau-
tions have been taken to shield the eyes from glare.
Semi-indirect illumination is quite feasible and the cost has
not been high. There are simple designs in pressed opalescent
glass bowls whch give excellent results, and the cost is very
slight. The use of white leaded glass semi-indirect units is a
very promising innovation, and it is possible to design units along
Gothic or Renaissance lines, so that they will harmonize ex-
cellently with the church architecture. I have in mind a church
in North Adams, Mass., where there are large fixtures of white
leaded glass of Gothic design 6 feet in diameter which hang
15 feet from the peak of the arch. Each fixture has 3 150-watt
and 3 250-watt tungsten filament lamps. Only a few built up
units of this type are necessary for a given space and conse-
quently the wiring cost may be reduced considerably.
This white glass has been utilized for direct lighting. An
asymmetrical reflector used in the Buffalo General Electric Com-
pany's building, was described* by Mr. Ryan, at last year's con-
vention. The design was slightly modified and the equipment
used for a church which had a fan-shaped roof that reached
a maximum height above the pulpit. Outlets were located on the
ribs and the fixtures hung to direct the maximum light toward the
front. There was sufficient diffused light transmitted through
the glass to light the balcony, and the main portion was lighted
by the reflected light.
The term "line of vision" is mentioned in the paper, and that
brings to mind the question: What is the line of vision? As a
homely illustration, one may note that the ladies of the church
often wonder why the gentlemen go to sleep so much more
readily than they do at a service. It has been my experience
that there is usually, within the angles of vision, a number of
bare or improperly shielded light sources, which shine directly
into my eyes. The ladies have their hats on while in church,
* Trans. I. E. S., Vol. VII, No. 8 (Nov., 1912), p. 597.
CHURCH LIGHTING 625
and hence, their eyes are more completely protected, and drowsi-
ness does not result as readily.
As regards the designing of special reflectors for chancel
lighting, I may say that we have had very good suc-
cess in lighting chancels of the arch type, by using angle
steel or glass reflectors, giving asymmetric curves. In the chan-
cel without the arch, there are quite often Corinthian columns
or similar structures on both sides, and by using the tubular type
lamp with a cylindrical or trough reflector, to direct the light to
the opposite wall rather than to the adjacent wall, as in cove light-
ing, good results have been obtained. When lighted from the
side rather than from overhead most chancels appear better
illuminated because the shadow effects are softened.
The lighting in the choir loft is often very annoying, for it is
usually accomplished by a number of small lamps with diffusing
shades which are in the line of vision. In some churches where
it has been impossible to install lamps within the interior of the
organ, or on the music racks, the opaque, bowl steel reflector,
with its exterior painted to harmonize with the woodwork, has
been used.
Window lighting from the exterior has been very satisfactorily
accomplished by the use of asymmetrical, weatherproof type steel
reflectors and regular lamps, or the tubular type lamp and small
reflectors arranged to evenly illuminate the entire window surface.
Prof. F. C. CaldweXl : The point was brought out that
the deck lighting as shown on Fig. 2, does not give suitable light-
ing for the upper part of the room. This is perhaps due to the
fact that the deck is somewhat recessed — the glass should be kept
down as near the ceiling as possible. In deck lighting it is impor-
tant that a glass of good transmission efficiency be used. Ordi-
nary skylight glass is very well for daylight, but seriously inter-
feres with the efficiency of a lighting system. Even where econ-
omy is not of prime importance the best results can usually be
obtained by putting money into good glass rather than into
additional power.
Mr. R. F. Pierce : There were one or two points brought out
in these two papers that are of especial interest in connection
with the paper presented yesterday by Dr. Ferree. There seems
626 TRANSACTIONS I. E. S. — PART II
to be a considerable movement in the direction of reducing the
brightness of surfaces within the range of vision. This is shown
by the popularity of direct and semi-indirect systems of lighting.
In churches, particularly, people are quite sensitive to the aggra-
vation produced by glaring light sources, and the fact that they
have often resorted to the so-called indirect system of lighting
is excellent evidence that our commercial glassware which, the
reflector manufacturer tells us effectively shades the lamp and
reduces the glare, really does comparatively little to that end.
The results given in Dr. Ferree's paper yesterday indicate that
reducing surface brightness of reflectors to the extent commonly
found in commercial types is not sufficient to make the surface
much less objectionable from a standpoint of depreciation in eye
efficiency than the bare lamp. We would, however, not be justi-
fied in concluding that indirect lighting as such is responsible for
the results obtained by Dr. Ferree. If we use a direct lighting
system in which the enclosing glassware forms the whole ceiling
of the room as shown in Fig. 2 in Mr. Ely's paper and in the
installation described by Mr. Kingsbury, we have an installation
in which the surface brightness of the glassware is no higher
than that of the ceiling in an indirect lighting system to produce
the same illumination, and we would certainly not expect any
different results as regards the efficiency of the eye. On the
other hand, if we use an indirect reflector concentrating the light
on a very small spot on the ceiling and producing a surface
brightness in the neighborhood of two or three candle-power
per square inch, we would have a condition that we would expect
to be quite as annoying and quite as unfavorable to the efficiency
of the eye as a direct lighting system in which a similar distri-
bution of surface brightness occurred. It appears that the prin-
cipal factor is the brightness and area of the illuminated sur-
faces, and it is immaterial whether a certain distribution is
obtained by "direct" or "indirect" means.
One serious drawback to the indirect system is the reversal
of the natural order of intensities. Under daylight, the higher
intensities are found at the lower levels, and the lower intensi-
ties at the upper levels. This condition is reversed with indirect
lighting on account of the fact that the reflecting surfaces are
CHURCH LIGHTING 627
generally diffusing, each element giving a circular distribution
curve, and it is practically impossible to avoid illuminating the
upper portions of the side walls to a higher degree than the lower
portions.
This effect may be avoided in deck lighting systems, however,
since the deck may be constructed of glass which, while suffi-
ciently diffusing to reduce surface brightness, will not seriously
interfere with the direction given to the light rays by the reflec-
tors. This effect is seen in the illustration of the deck lighting
systems shown by Mr. Ely.
Another important consideration in church lighting which fre-
quently militates against the employment of indirect lighting is
that of esthetic effect. The prevailing type of church architec-
ture is Gothic, in which it is the purpose of the architect to allow
the high, pointed arches to remain in comparative darkness.
When the indirect system of lighting is used, the brilliant illumi-
nation in this portion of the building entirely destroys the effect
which the architect strived to produce. Some particularly atro-
cious examples of a disregard for architectural considerations by
the application of indirect lighting to Gothic interiors have been
found in recent installations, and are excellent examples of what
good illumination should not be.
In the installation described by Mr. Kingsbury, use was made
of what might be termed a semi-indirect lighting system in
which the wall is used as a reflecting surface instead of the ceil-
ing. This is quite similar to one of the installations described
by Dr. Ives in reporting his experiments on the approximation
of daylight distribution in residence interiors. As I have not
had an opportunity to observe the results of lighting of this
character, I am unable to comment upon it ; but I think it presents
a problem worthy of more extended investigation.
Mr. J. R. Cravat 11 : In regard to Fig. 2 in Mr. Ely's paper,
I don't agree with Mr. Pierce that it represents an ideal condi-
tion, because just as Mr. Luckiesh has said, the contrast of the
brightness within the range of vision is the very important point.
In this case if we are to judge from the photograph (which,
however, may not represent things just as they are), we have
628 TRANSACTIONS I. E. S. — PART II
a very decided contrast between the brightness of the skylight
and the brightness of the ceiling and bright surroundings. Now,
that is probably about the best way that particular installation
could be lighted, but it does not illustrate an ideal condition,
because, that contrast must be annoying. Any kind of violent
contrast of surface brightness which one must face constantly
cannot fail to be annoying.
Mr. T. J. LiteE, Jr. : In a number of the installations de-
scribed by Mr. Ely, particularly the one shown in Fig. 2, it
would appear that mural decorations on the side-walls and ceil-
ings are not properly lighted. The decorations referred to
are usually very costly and in the lighting of such a building
the illumination must not only be sufficient for reading purposes
but must be of a character which will properly bring out the
decoration above referred to.
In reference to Fig. 1, I should say that a person sitting half-
way back in the church would see at least half of the fixtures.
In other words, the light sources would be within his range of
vision. I think the lighting of a church should be so arranged that
the lamps would be out of the range of sight. The light would
probably be in the speaker's eyes, but this could not be avoided.
At any rate, he is more apt to look down upon the congregation
and with the lamps hung high in the church they would not be
so apt to annoy him, and even if they did, from the standpoint
of the greatest good to the greatest number such an arrangement
would be considered preferable.
Why should the lighting of a church building be considered
so differently from the lighting of a theatre? In the latter case,
the lamps are of necessity shining in the player's eyes. Bare
lamps are never allowed to annoy the audience.
Mr. E. B. RowE: I have one or two questions to ask. One
Mr. Powell has touched on in connection with the use of semi-
indirect lighting. On the sixth page Mr. Ely mentions the fact
that semi-indirect lighting has been recommended where the condi-
tions were deemed very favorable. I would like as a matter of
information to have him give, if he can, what he considers the
conditions which are favorable to that system more than condi-
tions which are not.
CHURCH LIGHTING 629
In regard to the use of light units on the chancel side of
beams, etc., it occurred to me that there is one disadvantage in-
herent in that type of lighting and that is the glare in the eyes
of the pastor and the choir if it is on the chancel end of the
nave; perhaps he may have some information to give us as
to whether there have been any objections made from that
standpoint. That is an excellent method I think of obtaining an
efficient illumination very similar to direct lighting.
It occurred to me with regard to the lighting of croses on the
exterior of the church, which sometimes have to be located in
inaccessible points, that use might be made of the new con-
centrated filament incandescent lamp in parabolic reflectors, so
that there would be no need of ever going up to the cross itself.
Mr. LuckiESH : I want to supplement one of the points
brought out by Mr. Pierce by describing one of the most annoy-
ing cases of glare I ever experienced. This installation is in one
of the modern auditoriums in Cleveland which was installed by
an architect who I know has little use for an illuminating
engineer. The auditorium proper is lighted from beautiful in-
direct fixtures. The pulpit however is lighted by concealed
sources as in Fig. 2, of Mr. Ely's paper. When the pulpit alone
is lighted the contrast between its bright background and the
dark surroundings causes a most annoying glare. This brings
out the point that glare is not always due to high brightness.
Mr. G. H. Stickney: The author is to be congratulated on
having such a large number of successful church lighting installa-
tions. There is probably no other class of lighting problem
which the illuminating engineer approaches with more trepida-
tion, since the artistic requirement in church lighting is so pre-
dominant that the engineer, unless he can fully co-operate with
the architect, is at a tremendous disadvantage.
One of the most interesting problems in church lighting which
I have ever handled, and one which illustrates a novel method
which I have not seen used elsewhere, was in a large Gothic
cathedral in Montreal. This is a magnificent building erected
about 100 years ago. The main portion consists of three Gothic
naves, the center one being about 100 feet high. As the ceiling
is dark finish, the usual indirect lighting would have been unsuit-
63O TRANSACTIONS I. E. S. — PART II
able, although the equipment actually used might be classed as
indirect. In the previous installation fixtures were suported
from the backs of the seats and fairly brilliant light sources
located about 7 or 8 feet above the floor in such a manner as to
detract from the general view, especially of worshippers in the
rear seats. In the upper part of the building along each side
of the central nave was a line of windows which opened into a
covered space between the nave and the roof. Prismatic glass
was inserted in these windows and large tungsten filament lamps
arranged behind them, each equipped with a metal reflector so
as to direct the light through the window. The prism glass
deflected the light downward, and it was possible, by adjusting
the height of the lamps with regard to the windows, to control
the distribution of light and proportion it properly between the
upper and lower portions of the church. The energy consump-
tion was a little over one watt per square foot of floor area.
While I never had an opportunity of seeing the completed instal-
lation, it has been reported as producing a most pleasing effect,
and that it is possible to read in any part of the church.
Mr. R. B. Ely : In answer to Mr. Cravath's remark about
the installation in which the paintings appear — In that case we
used a cluster of tungsten lamps equipped with distributing type
reflectors of light opal glass to get diffusion, which was attached
to a rope and pulled up from the floor to the ceiling; it was
placed at various heights until we got that height at which the
paintings could be seen to best advantage. By increasing the
intensity (various sizes of lamps were tried) so as to get more
brilliancy all over the church, it was found that the paintings
could be seen more readily with a higher intensity than under
the lower intensity.
Mr. Powell touched on the intensity of illumination. It should
range, he stated, from ^4 to 1.5 foot-candles. However, some
churches, particularly Catholic churches, have to be lighted as
brilliantly as possible at certain times, Christmas and Easter for
example. Sufficient equipment should be installed to provide the
extra illumination required on such occasions.
Mr. Powell spoke about the question of lamps being in the
CHURCH LIGHTING 63 1
line of vision, and answered it himself. I think we can all tell
when a lamp is annoying or whether it is in the line of vision.
Reference was made to special reflectors for chancel illumina-
tion. There have been cases where it has been desirable to
illuminate the chancel more brilliantly than other sections of the
church, to bring out that part of the church. And in such cases
it is desirable to cut off the illumination from the lower and top
sides of fixtures to get that effect. We have in some instances
made special corrugated glass reflectors with definite cut-off points.
Reference has also been made to putting lamps behind Corin-
thian columns, with tubular lamps. We were called in on a case
that had such equipment, simply because the desired results
could not be obtained. The church panels which had been in
stalled at a cost of something like $2,000 were entirely flat with
this method of lighting; no irridescence from the tile panel was
to be had. Search-light lamps with parabolic reflectors were
used to direct light on these panels until the angle where the ir-
ridescence would appear to the congregation was found. Then
equipments were installed at those points, and the lighting was
found to be very effective.
In window lighting, as a rule, the main window or the most
beautiful window frequently appears at the rear of the church.
And the equipment for the gallery is generally located in that sec-
tion. With indirect and semi-indirect systems the fixture for this
portion of the church is usually located so that it will illuminate
that window to the best advantage, and at the same time provide
illumination for the gallery.
Regarding Mr. Caldwell's question about the introduction of
semi-indirect lighting — there are quite a number of cheap bowls
for semi-indirect light, but when one is dealing with a church
where generally very large units have to be used, I have invari-
ably found the cost to be greater than that of other types. It
doesn't make so much difference if it is a new installation, that
is in a new building being erected ; 'the cost is then not such a
factor. But if it is an old installation and there is competition
it is a factor.
Mr. Pierce referred to Fig. 2. The effect is not as it appears
in the photograph. There is considerable diffusion from the
6
632 TRANSACTIONS I. E- S. — PART II
walls which are very light buff; and the paneling of the ceiling
may be seen readily. It is not dark, as it appears in this photo-
graph; the darkness is probably due to long exposure in taking
the picture.
Mr. Luckiesh referred to the glare from the chancel. When
the rest of the lights in the auditorium are turned out we try
as far as possible to get a theatrical appearance, you might say;
that is, to concentrate the light on the pastor, as you would in
a theater concentrate the light on the actors. And frequently
where there is a very light background a portion of the lights
are turned out, excepting those that would tend to show the
pastor.
Mr. Litle commented on the same effect as produced in Fig. 2,
and as to reflection from walls, ceilings and decorations. These
are all well brought out under that installation. When planning
a lighting system for a church we consider the decorations. They
constitute a feature which should be properly illuminated.
I believe, as Mr. Marks said yesterday, that "any system of
illumination can be made very effective," and in all church instal-
lations we are largely governed by the architectural considera-
tions and the character of walls and ceilings.
EDWARDS AND HARRISON I ACCURACY OF PHOTOMETRY 633
SOME STUDIES IN ACCURACY OF PHOTOMETRY.*
BY EVAN J. EDWARDS AND WARD HARRISON.
Synopsis: Five separate investigations are reported in this paper:
I — Relative accuracy of Bunsen and Lummer-Brodhun devices. This test
involving several thousand readings showed the sensitivity, expressed in
average deviation, to be 0.4 per cent, for Lummer-Brodhun as compared
with 1.5 per cent, for the Bunsen. II — Magnitude of error due to parallel-
ism of rays, in the photometry of reflector sources. The results of tests
on typical reflectors for general illumination purposes show the errors to
be negligible. Ill — Method of investigating adjustment error; and the
calibration of portable photometers. This method consists in taking pho-
tometer readings for various distances from a working standard lamp and
analyzing the results by reducing the relation to a straight line function.
The constant of the photometer is obtained from the slope, and working
standard intensity. IV — Errors in illumination measurements due to fail-
ure of test plate to obey the cosine law. Discrepancies between measured
and calculated values of illumination are fully accounted for by this inves-
tigation. The average error of the plates investigated was found to be
over 10 per cent, at 450. Computations show that the photometer results
for an average installation are about 12 per cent. low. V — Method of
obtaining and recording distribution data. A so-called thousand lumen
basis of drawing distribution curves is proposed, in order to avoid error
and confusion in comparing reflector units. On this basis zonal lumen
values are instantly convertible to per cent, of total and intensity values
to per cent, of horizontal for the bare lamp. By using a single multiply-
ing factor all values can be corrected to current lamp efficiencies.
Many investigations which have to do with accuracy in photo-
metry have been made in the laboratories with which the authors
of this paper are identified. These investigations have served
their purpose as far as the particular laboratory is concerned,
but in some instances have not been reported. It is believed
that brief reviews of the more important ones will prove useful
to the members of the Illuminating Engineering Society.
This paper, then, is of the nature of a report on five separate
investigations.
* A paper read at the seventh annual convention of the Illuminating Engineering
Society, Pittsburgh, Pa., September 22-26, 1913.
The Illuminating Engineering Society is not responsible for the statements or
opinions advanced by contributors.
634 TRANSACTIONS I. E- S. — PART II
I — Relative accuracy of Bunsen and Lummer-Brodhun de-
vices.
II — Magnitude of error due to parallelism of rays, in the
photometry of reflector sources.
Ill — Method of investigating adjustment error; and the cali-
bration of portable photometers.
IV — Errors in illumination measurements due to failure of
test plate to obey the cosine law.
V. — Method of obtaining and recording distribution data.
No attempt is made to connect the various investigations, al-
though it will be seen that the method used in Investigation 2,
suggested the procedure of Investigation 3.
I. RELATIVE ACCURACY OF BUNSEN AND LUMMER-
BRODHUN DEVICES.
There is at present little question as to the order of sensitivity
of the common forms of photometric devices. The quantitative
results of an extended test to obtain relative accuracy values
of the Bunsen and Lummer-Brodhun devices, which resulted in
a decision to discard all Bunsen apparatus in favor of Lummer-
Brodhun, may, however, be of value.
About five thousand readings extending over a period of
three months were taken in order to establish beyond question
the relative accuracies of the two devices under the particular
conditions involved. The Bunsen photometer used was of
the regular type, a circular spot surrounded by a concentric field
and viewed by means of two angle mirrors. The Lummer-
Brodhum photometer used was of the low-contrast variety, which
has been shown* to be more sensitive than the earlier high con-
trast type. Except for the sight box, the same photometric
equipment was used for the entire test. Twenty incandescent
electric lamps of various kinds and efficiencies were used, all
against the same comparison lamp in order to get an idea as to
the effect of color difference.
The details of the test and many interesting but less important
deductions must be omitted. As an example may be cited the
test on the effect of changing the shape of the Bunsen spot.
It was found that a star shaped spot was easier to manipulate
* I^unimer and Brodhun, Zeitschrift fur Instrumenltnkunde, Vol. 9, p. 461.
EDWARDS AND HARRISON : ACCURACY OF PHOTOMETRY 635
and showed a higher sensibility although of the same order.
It is sufficient here to show the condensed results for the more
common type of Bunsen in the curves of Fig. I bearing in mind
that the precision is expressed in average deviation from the
mean and that ten readings were taken on each lamp with each
device for each set of observations. It is seen that the
Lummer-Brodhun precision is about four times as good as the
Bunsen. The actual grand average for the entire test is 0.4 per
cent, average deviation for the Lummer-Brodhun and 1.5 per
Fig. 1. — Relative accuracy of Bunsen and L,utnmer-Brodhun devices.
cent, for the Bunsen. No marked loss in precision results
from small color differences such as with carbon against tung-
sten.
II. MAGNITUDE OF ERROR DUE TO PARALLELISM OF RAYS,
IN THE PHOTOMETRY OF REFLECTOR SOURCES.
Many who have had occasion to calculate illumination from
the distribution curves of lamps with concentrating reflectors
have, no doubt, had a feeling of uncertainty as to the accuracy
of their results, due to the possible parallelism of a portion of
the light rays.
It would seem that where a reflector directs the rays to a
considerable degree, that there would be brought about an ap-
preciable effect of parallelism of the rays. This investigation
was undertaken with the idea of obtaining a measure of this
effect of parallelism in commercial types of reflector units.
In the theoretical extreme case of a parabolic reflector and a
point source, the illumination at various distances would be
636 TRANSACTIONS I. E. S. — PART II
constant. The point source with no reflector would, of course
furnish an illumination which would be strictly proportional to
the reciprocal of the square of the distance. A partial parallel-
ism, such as results from the use of a directive reflector, would
be expected to bring an illumination which does not decrease
with distance as much as would be given by the inverse square
law.
In order to determine the magnitude of these effects, tests
were conducted on a number of typical reflectors using a Weber
portable photometer in such a manner as to eliminate other
sources of error. The method employed depends upon the
principle that when a photometer screen is balanced between
two point sources of constant intensity the ratio of the distances
of the two lamps from the screen is constant. Readings were
first taken on a bare lamp at distances varying from 3 to 25 feet
0.914 to 7.315 m.) to serve as a test of the adjustment of the
photometer, and then the bare lamp was replaced by various
reflector units in turn, and readings obtained. The relation
between the test and comparison lamp distances from the photo-
metric screens is a straight line for point sources, and moreover
a straight line passing through the origin. Therefore a plot
of distance of the test unit against photometer reading (since
a Weber photometer reads comparison lamp distance directly)
serves as a complete test of the inverse square law is applied
to the unit in question.
The distance of the light unit was varied by raising and lower-
ing it, the portable photometer being set up directly beneath.
Readings were taken in the direction of the axis of symmetry of
the unit, since the effect should be most pronounced at this
angle.
The results for several types of reflector units are summarized
in the curves of Fig. 2. They are self-explanatory, both as re-
gards quantitative results and their precision since the actual
points are shown. Each point is obtained from the average
of four photometer readings. The abscissae of the bare lamp,
extensive and intensive curves, have been multiplied by two in
order to show all curves conveniently on one sheet. Therefore,
EDWARDS AND HARRISON '. ACCURACY OF PHOTOMETRY 637
the intercepts of the extensive and intensive curves, as drawn,
show twice their actual values.
It is seen that the curves are fairly good straight lines and
also that the intercept is within the body of the filament in
every case.
A failure of the line to pass through the origin shows that
the distance which should be used in computing illumination by
the inverse square law may not be exactly the same as the dis-
tance to the center of gravity of the filament. It appears from
the curves that such a discrepancy is more likely to occur than
?/
-
/
f
r
i
-
v
i
j
^
,--
•
•"
\
I
A
1
y
-y
<■-
-
*
>'
-
9
>
>
t
-
-
/
y
t
/
•
-
--
:
l =y
Si
^
i
.-•
-
*
--'
31
S£
I %
N
/
t
r
-
r
y
£
9
tf»
it
.-
*
J
•-.'.
J«
IM
i£rv
?.S
y
t
•*
*$
n
mz
m
oJ
■
/
•
>
ta
a
r-7_
.tlis
'■
a
■ ■
::-'.
-i .
■ c
« 1
-
'■ -
" _
;".
--. !
-V-
';;
,.■=
M
r
'
c,- j
Fig. 2. — Curves of photometric readings and distances for reflectors, showing relative
intercepts. Data obtained with 100-watt tungsten lamp. (The slopes of the bare
lamp, extensive and intensive curves, have been changed by multiplying their abscis-
sae by 2).
an error due to a failure of the inverse square law itself. This
is shown by all the curves being straight lines having slightly
different intercepts. There may be an appreciable shifting of
the effective luminous center, but there is no appreciable devia-
tion from the inverse square law.
Errors in illumination calculations due to an error in the dis-
tance used in computations is given by curves of Fig. 3. The
error due to assuming the distance — that to the center of gravity
of the light source — is probably less than 1 or 2 per cent, in the
usual case where the unit has been photometered at the distances
of about 10 or 12 feet (3.048 or 3.657 m.).
The error due to parallelism must be small, as will be seen by
638
TRANSACTIONS I. E. S. PART II
reference to curves of Fig. 4. These show the effect on the
graph when certain percentages of the illumination at 10 feet
(3.048 m.) are assumed to be the result of parallel rays. The
deviation from a straight line becomes very marked.
8
z
27
-5.
o;Z6
•-""4
zo
i-UUJ
0P3
K?
0
*-»
1
0
<?
x^
V*s»
k*
^K.
2£s>
S£/cg-
s^
3 Z 4 6 8 10 12 14 16 18 20 22 24 26 28 30
DISTANCE, FEET
Fige. 3. — Curve showing errors in illumination computations resulting
from error in the distance.
26
*W
on
£22
UJ
|20
§18
<n
§16
§K
UJ
CC
ot2
5
£8
0
£6
4
2
0
CUF
IVE
A-
w\
fyp
<f°
tf$
tfl°
) 2 4 6 8 10 12 14 16 18 20 22 24 26 28 50
DISTANCE, FEET
Fig. 4. — Effect of parallel rays on measurements of illumination. Curve A is one which
would be obtained from the point source ; other curves show effect of parallel rays
for given percentages of illumination at 10 ft. distance, due to parallel component.
The results of this investigation seem to justify the conclu-
sion that for all practical purposes illumination from general
lighting units may be computed by using the inverse square
EDWARDS AND HARRISON : ACCURACY OF PHOTOMETRY 639
law and taking the distance as that to the center of gravity of
the light source. It is unnecessary to specify the distance at
which distribution curves are taken, provided, of course, that
they are taken at a distance several times the greatest dimension
of the unit. Also, it is unnecessary to use the term apparent
candle-power in connection with the directional intensity of
such reflector units as are used for ordinary illumination pur-
poses.
III. METHOD OF INVESTIGATING ADJUSTMENT ERROR, AND
THE CALIBRATION OF PORTABLE PHOTOMETERS.
The method used in the investigation of the effect of parallel-
ism in the photometry of reflector units suggested itself as
being a very good one to apply in the calibration of portable
photometers. In fact, in the previous work a portable photo-
meter was used as the photometric device, and the bare lamp
test originally showed the photometer to be out of adjustment.
The adjustment error was corrected by means of the data ob-
tained on this preliminary run.
Portable photometers have a variety of means of varying the
brightness of a comparison surface which is matched with the
surface illuminated from the test end. The general method, here
described, is applicable to all when consideration is given to the
particular principle by which they work.
Consider first the type where the distance from the compari-
son lamp to the diffusing plate is varied, such as the regular
Weber photometer.
Let I1 = c-p. of a working standard which can be placed any
distance from screen.
I2 = c-p. of comparison lamp.
dx = Distance of It from its diffusing plate.
d„ = Distance of I2 from its diffusing plate.
When a photometric balance is obtained
K } d? ' d?
where /t and /, are respectively the transmission coefficients of
the diffusing plates for lx and Ir
640
TRANSACTIONS I. E. S. — PART II
— is a constant and therefore may be placed equal to Kj or,
Since the scale of the Weber is graduated to give d2 directly,
the reading
(3) R = ^m'tt-
Considering Ix and I2 constant, the relation of R and d1 is, of
course, the equation of a straight line passing through the
origin.
26
111
u-22
cZ
0
g|2
u-10
0
UJ 0
O 0
Z
S6
54
2
0
■ -/
■"/
//
A
//
mi
> y
N0.1
//
A
4t
/f
Vi
' 2 4 6 8 10 12 rt 16 18 20 22 24 26 28 30
PHOTOMETRIC READINGS, CENTIMETER5
Fig. 5. — Adjustment and calibration of Weber photometric.
It is seen from equation 3 that a test run made by setting up
a working standard lamp at various distances from the screen of
a portable photometer serves as a complete test of the accuracy
of adjustment and of scale graduation. A curved or irregular
line would indicate errors such as incorrect graduation of the
scale. The failure of the graph to pass through the origin
indicates an error such as incorrect position of the comparison
lamp. Curve 1 of Fig. 5 illustrates the manner in which in-
correct position of the comparison lamp is shown by this method.
EDWARDS AND HARRISON I ACCURACY OF PHOTOMETRY 64I
Curve 2 of Fig. 5 shows the results, as obtained, after correcting
the comparison lamp position by measuring the intercept of
Curve 1. It is interesting to note in this connection that the
actual position required to give the zero intercept does not cor-
respond exactly with the measured value, due, probably, to the
reflection of light on the inner surface of the tube.
Letting C represent the multiplying factor, which must be
applied to the reciprocal of the square of the photometer read-
ings to obtain illumination values in foot-candles, it is seen
(R\ 2
— ) , and
1
it follows that the constant of the photometer is given by
multiplying the intensity of the working standard by the square
of the slope of the graph.
For a photometer, such as the Sharp-Millar, where the read-
ings are made proportional to the reciprocal of the square of
the distance to the comparison lamp, the equations are changed.
Here the relation of d. and . ■ should be a straight line pass-
y R
ing through the origin. Since the slope of the line in this case is
equal to , the constant of the photometer is given by mul-
tiplying the intensity of the working standard by the square of
the slope. A photometer of this kind is usually calibrated to
have unity constant, so this method does not lend itself particu-
larly well for purposes of calibration, but is very useful in
testing out the adjustment of the instrument as well as the
accuracy of a particular calibration. In the laboratories with
which the authors are connected, special equipment is provided
and tests of this kind are applied to all portable photometers at
frequent intervals.
The change of constants effected by the use of absorbing
screens mav also be accurately determined once for all bv this
method. If there are a number of absorbing screens which have
transmission coefficients mlt tn2, etc., which can be used either
on the test or the comparison end of the photometer, the new
642
TRANSACTIONS I. E. S. PART II
constant C is equal to — if a screen is used on the test end
7)1
and Cm if used on the comparison end. The change in con-
stant is then obtained by a calibration test with and without an
absorbing screen, and the transmission coefficient is given by
the ratio of the two constants thus obtained. It will, of course,
remain unchanged except as affected by collection of dust or
dirt.
IV. ERRORS IN ILLUMINATION MEASUREMENTS DUE TO
FAILURE OF TEST PLATE TO OBEY THE COSINE LAW.
In measuring the illumination in the usual lighting installation
by means of a portable photometer, the light reaches the test
plate at practically all angles. A test plate which fails to obey
S.9
z.2
i=0
1 1
[_
-CA
LCI
LAI
ED
FTOM
COS
1NE
I LAW
1
te;
T F
ESI
Lli
1
v-
iv
PEF
<CE
NT
ERF
!0R
70
60
50,
40 uj
10
0 10 20 30 40 50 60 70 80 90
DEGREES FROM NADIR
Fig. 6.— Error in photometric test plates, due to failure to conform to the consine law.
the cosine law will give proper weight to only the light which
strikes it at an angle normal to it, assuming, of course, that the
photometer has been calibrated in the normal. This investiga-
tion was begun after noting discrepancies between measured and
calculated illumination values for several installations. The dis-
crepancies were noted particularly where extensive-type reflectors
were used.
It was rather surprising to learn that no flat plate could be
found which did not show a very considerable error. It was also
rather surprising to find that all the plates tested, although ob-
EDWARDS AND HARRISON : ACCURACY OF PHOTOMETRY 643
tained from different sources checked within i or 2 per cent.
The complete results together with a curve showing the per cent,
error are given in Fig. 6. The illumination values are given for
convenience in terms of ratio to the normal.
H
Era
E
X
2
|
"2.0 1.8 1.6 1.4 L2 LO tt8 0.6 Q4
SPACING RATiaRATI0 = -j^j|^7
Fig. 7.— Average per cent, error in illumination values for an installation of a large
number of units. (Extensive enamelled steel type on 10 ft. centers.)
35° 25° 15° 5° 0 5° 15° 25° 35°
Fig. S. — Photometric distribution curve of bowl shaped enamelled steel reflector.
[—16
W
gtz
cc
£8
36
a:
1114
2
2
0 L8 L6 1.4 1.2 1.0 08 0.6 0.4
SPACING RATIO.RATIO'^g1^
Fig. 9. — Average per cent, error in illumination values, for an installation of a large
number of units. (12. in. mirrored reflectors on 10 ft. centers.
It is seen that the error at 450 is more than 10 per cent. The
curves of Figs. 7, 9 and 11 are the result of laborous computa-
tions and are given as illustrations of the magnitude of the errors
which are obtained with commonly used systems in large rooms.
Fig. 7 is an example of extensive distribution being for bowl-
644
TRANSACTIONS I. D. S. — PART II
shaped enamel steel having the distribution shown in Fig. 8.
Likewise, Fig. 9 is for the narrow distribution shown in Fig. 10.
Fig. 11 is still another example showing the error for another
type of extensive distribution shown in Fig. 12.
25* 15" 5" 0 5" IB-
Fig. io.— Photometric distribution curve and mirrored reflector.
[— \b
14
gl2
gio
UJ
<J6
a;
£4
Z
°2
i
0 L8 L6 M LZ LO 0.8 0.6 0.
SPACING RATiaRATO^PACING
4
Fig. ii.— Average percent error in illumination values, anfor installation of a large
number of units. (Fnamelled steel dome shaped reflectors on io ft. centers.)
Fig. 12.— Photometric distribution curve of enamelled steel dome shaped reflector
The results of this investigation point to the conclusion that
the error in making illumination readings with the average flat
test plate is more than io per cent, in the majority of cases. The
discrepancy between measured and calculated values which have
EDWARDS AND HARRISON '. ACCURACY OF PHOTOMETRY 645
been observed are fully accounted for. For accurate work a care-
fully designed curved surface test plate should be used, or the
proper correction should be applied where flat plates are used.
V. METHOD OF OBTAINING AND RECORDING
DISTRIBUTION DATA.
Distribution curves on various reflector units are largely for
LAMP NO- Q.W-L0
VOLTAGE! K 3VLB£lear-J-30JL
TOTAL LUMENS JQOO.
LUMENS PER WATT- !0-
RUDUCTION FACTOR £W_.
REFLECTOR
DIAM..AL" DEPTH- 7.
HOLDER _ 3= tinker.
TEST ON
TESTED BYi CHECKED BY....
APPROVED BY. -JiCALi
PATE _ & il-il- ___ _ -
REMARKS SjroleCujyeJJniTS^Sini.
ANGLE
ZONE
LUMEN!
0
:: • 5
5
m-.s
0 10
17.0
15
185-6
0 20
09.5
25
mo
0 30
1635
35
ms
to
?WS
45
less
0 50
ss
1059
0 56
5Si-5
65
1404
0 70
»«
75
66-5
0 60
c'4o
85
70
0 90
Sii-S
So
is
0 100
105
0 MO
115
0 IZO
125
0 1.30
135
0 iiCL
1451
0 150
155
0 160
165
0 170
'175"
0 180
180
c,r
ZONE
LUMENS
v,^
W
060
J83.5
S8-4
c-,0
■i"
O-90
Sir 5
szt
ISiO
r: > ifir
0.0
O.t
0-0
*
K1J-160
00
Oi
6 0
0-!60
ez'.s
su
es3
TO MAINTAIN UNIFORMITY IN
REC0RD1N6 DATA THE LAMP USEPIN
THISTEST WAS OPERATED AT AN EFFI.
aft t C1ENCY OF 10 LUMENS PER WATTOR...
A W.RCTOREDUCETO W.RC
' MULTIPLY CANDLE-POWER AND
LUMEN VALUES SC
Clear 100 Watt lamp
Fig. 13-
purposes of comparison in order to determine the most suitable
for a case in question. The fact that these curves have been made
646 TRANSACTIONS I. E. S. — PART II
with lamps operated at various efficiencies and with different
reduction factors has resulted in considerable confusion and many
erroneous conclusions. The practise has been to obtain and put
on file a curve for each kind and size of unit, and to revise the
entire sheet each time thevlamp efficiency changes.
A new method, where the curves are obtained by using a lamp
having a known and standard reduction factor and operated at
an efficiency of 10 lumens-per-watt, has been devised. A sample
curve is shown by Fig. 13. It is intended as a curve which is
representative of a type of reflector and which can be applied to
various sizes and with lamps of various efficiencies. It will be
noted that the values are on a basis of 1,000 total lumens for the
lamp, which is very nearly correct for the 100-watt lamp at 1
watt per horizontal candle-power. Assuming the standard reduc-
tion factor 78 per cent, for the 100-watt straight-sides lamp, the
total lumens at 1 watt-per-candle are 980, or 2 per cent, less than
the 1,000 used. On the basis of a reduction factor of 81 per cent.,
which is standard for the larger round bulb lamps, the total
lumens for 1 watt per horizontal candle-power would be 1,018.
Therefore, the specific consumption is within 2 per cent, of 1
watt-per-candle for both the straight-sides and round bulb lamps
on the basis of 10 lumens per watt, and the average happens to
be within 0.1 per cent.
It is seen that all values of lumens are instantly convertible to
per cent, of total. The intensity values are as easily converted
into per cent, of the horizontal for the bare lamp, with an error
not greater than 2 per cent. The original sheet is provided with
a blank space for the multiplying factor which must be applied
in order to change to a specified efficiency, the intention being to
fill in, on blue print copies, the proper factor at the time the
sheet is sent out. Where a typical curve is drawn for a type of
reflector for all of its sizes, the data for most purposes can be
corrected by multiplying by the ratio of the wattage in question
to 100 watts. A further correction for efficiency can be made
where high accuracy is desired.
For purposes of comparing reflectors no reductions are neces-
sary. The comparisons can be made quickly and without danger
of failing to take into account the differences in the lamps used.
EDWARDS AND HARRISON \ ACCURACY OF PHOTOMETRY 64/
It has been found that the number of curves necessary for
ordinary illumination work can be cut down to a small part of
that formerly used, and, as lamp efficiencies change, it is only
necessary to supply a new multiplying factor to bring the curve
sheets up to date.
DISCUSSION.
Mr. T. H. Amrine: With reference to the comparisons of
the Bunsen and Lummer-Brodhun photometer. I should like to
ask Mr. Edwards something about the experience of the operators
in this test with the Bunsen photometer as compared with
the Lummer-Brodhun, whether they had equal experience on the
two photometers. The reason I ask this is that we get on the Bun-
sen photometer with experienced photometrists somewhat better
accuracy than that indicated by an average deviation of 1.5 per
cent. For instance, I have the result of tests of 3 different
operators on both the round and the star shaped spots from
some 20 tests, in which the photometer bar has masked so that
the operators could not see what readings they were getting.
The average deviation for the 3 operators was 0.497 per cent.
These were, of course, experienced photometer operators.
Mr. S. L. E. Rose: The paper just given by Mr. Edwards
contains a great deal of valuable information to photometrists.
The part of the paper I want to emphasize is that of making
distribution curves using a 100-watt lamp operated so as to give
1,000 total lumens. This one curve can then be used for other
reflectors of the same type and different size lamps by simply
applying a factor. The factor will also take care of the effici-
ency at which the lamps are to be operated in practise. This
will probably not appeal to the layman as usually he does not
want to bother multiplying by any factors ; but to the engineer
it should appeal very strongly.
Mr. L. J. Lewinson : Some of us are probably interested
in the commercial aspect of photometry. It might be in order
to ask Mr. Edwards about the relative speed attained with the
Lummer-Brodhun and Bunsen photometers. Presumably the
time reuired to make a measurement is longer with the former.
Does the increased accuracy of the Lummer-Brodhun photo-
7
648 TRANSACTIONS I. E. S. — PART II
meter overbalance the extra time required for a measurement,
and also the increased eye fatigue which probably results?
While I have no figures which are directly comparable to the
results shown in Fig. I, the following records may be of interest:
At the Electrical Testing Laboratories we use 3 photometers
equipped with Bunsen screens for our commercial work. As a
regular routine, about 20 per cent, of the lamps measured at
each photometer are checked at another photometer. One
photometer is considered as the standard, and the variations of
the measurements made at the other photometers computed.
We have made daily averages of these variations over long
periods, and find that during one year the average variation
from the mean value was about 1.17 per cent, and for the follow-
ing year about 1.2 per cent.
Dr. H. E. Ives : In connection with this discussion, several
years ago a paper came out by Dr. A. E. Kennelly in the Trans-
actions, of the National Electric Light Association entitled
"Accuracy in Photometry."* I would recommend that those in
terested in the present paper to look it up. Several of the ques-
tions named to-day are answered in it.
Mr. M. LuckiEsh : This paper has brought out some par-
ticularly interesting points and the measurements have settled
some doubts in my mind in a graphical manner. In regard to dif-
fusing screens, the best that I could make or otherwise procure
showed errors of about the same magnitude as those examined by
the authors. Owing to the failure of the screens to be perfectly
mat we might be led to believe that this is always a serious
matter. In a room with dark walls the illuminometer measure-
ments will be smaller in value than the computed measure-
ments. However in all practical cases there will be some light
reflected to the photometer screen by an indirect route. This
will tend to compensate the apparent deficiency. In many in-
teriors this additional light flux which is not computed will
amount to an appreciable fraction of the total light flux falling
on the working plane. This fault of diffusing screens is there-
fore not a very serious matter in most cases. The authors
have illustrated some simple procedures which might well be
* Proc, National Electric X,ight Association (1908), p. 208.
ACCURACY OF PHOTOMETRY 649
adopted by those wishing exact knowledge of the accuracy of
their photometric data.
Mr. E. J. Edwards (In reply) : Regarding the question
brought up by one speaker, I would say that the inverse square
law applies even to very concentrating headlights at distances
beyond a certain point up to which there is likely to be crossing
of the rays from the unit. Attempts to carry on investigations
along the lines of this test with such highly concentrated units
as ordinary headlights are very difficult, due to lack of uniformity
of the cross section of the beam. It is very hard to place the
photometer to catch the same beam at the various distances.
Tests, so far as they have been carried on, seem to indicate that
the inverse square law holds even in the case of headlights and
other projecting units, although the effective distance as obtained
by a graph might not show as good agreement with the actual
measured distance as in the case of the less concentrating type
of units used for general illumination purposes.
Mr. Luckiesh asked as to the effect of the diffused light on
the test plate error curves in any assumed system. In the com-
putations shown, equally spaced stations were taken and the
illumination computed from a distribution curve and the summa-
tion was carried out to a distance away from the unit where the
increment of illumination became negligible. This method takes
no account of the diffused light from walls and ceiling. I think
it would be very hard to determine accurately just how much
the error curve would be affected by diffusion, but surely, in
the cases shown, 90 per cent, or more of the total light reaching
the test plate would, under average conditions, come directly
from the units, and therefore, the effect on the curve would have
to be very small. Even the diffused light component would have
to have somewhat the same error since it reaches the test plate
from all angles.
It is gratifying to the authors to note that Mr. Rose empha-
sized the importance of considering the general adoption of a
simplified means of handling distribution curves such as the
thousand-lumen method suggested in this paper. In the case of
a curve applying specificially to a 250-watt unit, rather than to
65O TRANSACTIONS I. E. S. — PART II
a type of reflector, it would probably be better to plot on the
basis of 10 lumens-per-watt instead of 1,000 total lumens.
In reply to Mr. Lewinson's question regarding eye fatigue and
the time taken to make settings with the Lummer-Brodhun pho-
tometer, I would relate briefly our experience in changing over
our equipment. When certain operators, who had been accus-
tomed to using the Bunsen screen, were asked to use the
Lummer-Brodhun, they complained very much because of the
fatigue which seemed to result from using one eye instead of
two, and also because it seemed to take longer for settings. It
was only by promising these operators that they would not have
to use the Lummer-Brodhun unless they preferred them at the
end of a trial period that we were able to get them to willingly
try them out. At the end of a week they said they were able
to make the settings more easily and more quickly, although
they used but one eye. The narrower range of doubt in the
settings seemed to give the operators a greatly increased con-
fidence in their work.
In this connection I might also answer Mr. Amrine's question
regarding the experience of operators. The results as shown in
Fig. 1 were for an operator about equally experienced with
both photometers. There were other results obtained in this test
by operators who had been accustomed to using only the Bunsen
screen, but since a fair comparison of inherent accuracy of the
two systems was desired rather than the accuracy as applied to
any certain class of operator, they are not included in Fig. 1.
It was found that the operators who were unaccustomed to the
Lummer-Brodhun read with nearly as low an average deviation
as the experienced operator, but were incorrect in actual value be-
cause they set according to some appearance of the field other
than that resulting from equal intensity on both sides. I agree
with Mr. Amrine that it is possible for an experienced Bunsen
operator to obtain average deviations as low as ^ per cent, as
noted by him. But on the same basis we believe a very expert
operator under most favorable conditions might show less than
0.2 per cent, with the Lummer-Brodhun.
The flicker photometer was not considered at all, since it had
never been the subject of thorough investigation by us. The
ACCURACY OF PHOTOMETRY 65 1
paper is not intended as a comprehensive one on the subject of
photometry.
Another speaker has shown a test plate error curve plotted
on the same basis as our Fig. 6. It is interesting to note that
his tests show a smaller error. Our test included some plates of
the same make as those he mentioned, and, although they showed
a somewhat less error than the average, they came out greater
than the result he showed. All of the plates tested showed the
same characteristic shape of error curve. We did not find that
the error reached a maximum at a certain intermediate angle. It
seemed to increase in a smooth curve out to the limiting angle
where the error value becomes indeterminate. Even with plates
as good as he used it is evident that test plate error is important,
and accounts for many discrepancies noted in the past.
652 TRANSACTIONS I. E. S. — PART II
THE STATUS OF THE LIGHTING ART*
In every branch of human endeavor there comes a time when
it is advisable to appraise progress. The periodic test of stu-
dents' knowledge, the counting of cash in the till, the taking of
stock in a mercantile establishment, the mariner's observation,
reconnaissance in military manoeuvres, — are all recognitions of
the need for measuring progress and determining status. In all
forward movements such determinations are admittedly indis-
pensable.
The science and art of illumination have made considerable
progress in the recent past. The growth and improvement have
been rapid and of a nature to command widespread attention
from scientific, commercial and humanitarian viewpoints. Ram-
ifications are so numerous, however, and the variation in practise
is so great that, even in the opinions of those best qualified to
judge, it is difficult to fix the present status.
To possess a thorough knowledge of the status of the lighting
art would be of great advantage. First, it would constitute a
record for future reference and for comparison with corres-
ponding records which may be compiled in later years, and,
second, it would indicate points of greatest weakness in knowl-
edge and practise, and make apparent the directions in which,
because of such weakness, this Society has the greatest oppor-
tunity for useful service in the near future.
Because of conviction that a very useful purpose would be
served by a record of knowledge of the principles of illumina-
tion already attained, and by a review of existing practise in
the art of illumination, it has been thought advisable to under-
take the compilation of facts in an endeavor to record the status
of the lighting art in the United States of America in the year
191 3. The obstacles in the way of success in such an effort
are, however, considerable. The range of variation in practise
is so large as to make it impracticable to present a really com-
prehensive survey. More -numerous agencies than are available
are required in order to obtain sufficiently reliable and consistent
* Presidential address by Preston S. Millar at seventh annual convention of Illumin-
ating Engineering Society, Pittsburgh, Pa., September 22-26, 1913.
THE STATUS OF THE LIGHTING ART 653
records of practise in various parts of the country and in the
several branches of illumination. To present a fittingly au-
thoritative review, a more accurate and detailed knowledge than
the writer possesses is essential. These obvious obstacles, how-
ever, were deterrents of too mild a nature to dissuade from an
attempt regarded as desirable.
With a view to securing information regarding lighting prac-
tise, questions were prepared and issued about the first of July
to representative individuals and commercial organizations of the
following classes :
Central stations (300 lists).
Gas Companies (300 lists).
Municipal engineers (200 lists).
Ophthalmologists (200 lists).
School associations and commissions (200 lists).
Street railroad companies (200 lists).
Manufacturers of incandescent lamps (in various quan-
tities).
Manufacturers of mantle burner lamps (in various quan-
tities).
Manufacturers of arc lamps (in various quantities).
Manufacturers of acetylene supplies, tips, etc. (in various
quantities).
Manufacturers of oil lamps (in various quantities).
Manufacturers of small isolated lighting plant equip-
ments (in various quantities).
Manufacturers of gasoline, acetylene, etc. (in various
quantities).
Manufacturers of lighting glassware (in various quanti-
ties).
Fixture manufacturers (in various quantities).
Arc lamp post manufacturers (in various quantities).
Railroads (in various quantities).
Street lighting companies (in- various quantities).
The country was divided into areas of approximately 1,000,000
population each, and so far as practicable, lists of questions
were sent to possible sources of information in each section of
1,000,000 population. In all there were 128 questions, many being
654 TRANSACTIONS I. E. S. — PART II
duplicates of questions which appeared on other sheets. Approxi-
mately 1,750 lists of questions were issued. The questions as a
whole dealt with the recognized fundamentals of illuminating
practise, though in all cases the attempt was made to adapt them
to the knowledge which the correspondent was understood to
possess.
It was anticipated that the replies to these questions would be
relatively few in number and that the information thus fur-
nished would be inadequate, both in respect to reliability and
comprehensiveness. It was felt, however, that in the aggregate
the information elicited would be interesting and valuable.
Furthermore, the submission of these questions would be of some
value in attracting attention to the need for improvement in
lighting practise, and in drawing attention to this Society as the
exponent of good illumination.
Approximately 20 per cent, of those receiving the questions
promised to supply such information as they possessed or could
obtain, and approximately one-half of these have been heard
from.
In discussing the subject, it is necessary to adopt some form
of classification, if each department of knowledge and practise
is to be considered intelligently. It is possible, and it is perhaps
customary, to classify with respect to the nature of the installa-
tion and of the premises illuminated. It has seemed preferable
in this case to classify according to those features which deter-
mine the success or the failure of the illumination, irrespective of
the nature of the installation. Light should be of proper quality
and the lighting equipment should be installed suitably. In the
present stage of our knowledge it is considered that illumination
is satisfactory if it is correct in regard to certain qualities of
light and to certain features of utilization. These it is presumed
to present to you as —
The Categories of Illumination.
Light Utilization
Intensity Contrast
Direction Cougruity
Diffusion Hygiene and Safety
Color Cost
Steadiness
THE STATUS OF THE LIGHTING ART 655
The categories are believed to be inclusive of all ordinary
features of illumination which should receive attention from the
illuminating engineer. They are not independent of one another,
but are interconnected in such a way as to make it impracticable
to discuss any one without referring to one or more of the others.
It is also impracticable to arrange them in order of importance,
since their relative importance varies with local conditions and
with the requirements of the installation.
In the following discussions of the various qualities of light,
features of lighting practise and correlated matters, the informa-
tion made available through the lighting survey which has been
described has been combined with such material as could be
obtained from other available sources, and the whole is com-
pressed into a brief review.
INTENSITY OF LIGHT.
Intensity is a quality of illumination which has received very
general attention from the earliest days of illuminating engineer-
ing. Its great importance was recognized early, and led to
study and to the development of photometers for facilitating such
study. Early writings on illumination rarely failed to include
a table setting forth views as to the intensities which ought to
prevail in various classes of installations. It is true that these
statements were very largely restricted to the mean horizontal
illumination at some height, usually 30 inches, above the floor,
and neglected other important aspects such as wall brightness.
Yet emphasis upon intensity undoubtedly prompted increase of
light in installations where it had been inadequate, and tended to
raise the standard everywhere.
In many classes of lighting the intensity standard has been
increased greatly within the last few years. Street lighting, very
generally inadequate, has felt this advance. The use of more
powerful illuminants, the growing appreciation of importance of
lighting business streets well, and' the influence of merchants'
display lighting systems have operated to increase the standard of
illumination intensity. In the middle and better class stores a
high intensity of illumination has been found to have a merchan-
dising value, and the intensity standard has been largely increased
in the past few years. In large intelligently conducted industrial
656 TRANSACTIONS I. E. S. — PART II
plants the influence of improved lighting upon output and upon
safety of employees is evidenced in increased light intensity. In
small factories, as in small stores, this improvement is less
marked. Sign lighting in the past few years has increased greatly,
indirectly promoting intensity increase particularly in street light-
ing. In residences the advance has been felt perhaps less than
elsewhere.
The attitude of manufacturers of illuminants and of lighting
companies is an important element in determining the trend of
practise in regard to intensities. Manufacturers of gas mantle
lamps have been active in promoting the use of their product,
displacing open flame burners with large increases in intensity.
It would appear that there exists to-day an opportunity for gas
companies to contribute largely to the improvement of illumina-
tion by more actively promoting the substitution of mantle lamps
wherever open flame burners are still in use.
In the electric lamp field the manufacturers have exploited the
Mazda lamp widely and their efforts have been seconded more or
less by central station companies. In some cities the central
station company has pushed the use of the Mazda lamp actively
with very beneficial results. In other cities the central station
company has not taken an active part in making this improvement
in electric lighting available to its customers. An indication of
the extent to which the Mazda lamp is now employed will be
found in the sales records of the electric lamp manufacturers,
which show that for the 3 months May, June and July, 191 3,
the sales of the several types of lamps have been as follows :
Per cent.
Gem and Carbon 37
Mazda 63
When it is remembered that each Mazda lamp lasts two or three
times as long as a carbon or Gem lamp, it will be seen that the
use of Mazda lamps has now become very general, especially
when it is recognized that in general it is the more extensively
used lamps which are of the Mazda type.
The sales records of lamp manufacturers furnish another inter-
esting evidence bearing upon the increase in the light intensity
standard. This is the average candle-power of the lamps sold
during recent years.* These records are as follows :
* Courtesy of General Electric Co.
THE STATUS OF THE LIGHTING ART 657
Approximate average candle-
power of all incandescent
Year lamps sold
1906 18
1907 19
1908 21
1909 23
1910 25
1911 26
1912 29
1913 (estimated) 32
Standard practise in regard to light intensity depends upon
conditions other than simple illumination requirements. The
standards vary in installations of similar class in different cities.
For example, in some cities merchants have not come to appre-
ciate the advantage of adequate, well designed illumination to
the extent that they have in other cities. Again in some cities
the installation of "white way lighting" has been found to mili-
tate against successful show window lighting, the street light of
higher intensity being considered sufficient to illuminate the show
windows for ordinary purposes. In other cities similar installa-
tions have operated to increase the intensities employed in show
window lighting, it being found that in contrast to the lighting of
the street, more light is necessary in order to make the window
displays as prominent as they were when the street lighting was
inadequate.
In few classes of installations is it practicable to measure the
advance in light intensity throughout a period of years. While
representative data are available for modern installations, very
generally they are lacking for the older installations. In railway
car lighting there is an opportunity to measure the progress
which has been made, due to the fact that the older cars, which
have antiquated lighting systems, are continued in service on
branch lines long after their type has become obsolete. Minick*
has availed himself of this opportunity to place upon record the
average illumination intensities which are typical of the several
systems of lighting which have been used in the lighting of day
coaches. This record is as follows :
* I. E. S. Transactions, May, 1913, page 214 and communication to the writer.
658 TRANSACTIONS I. E. S. — PART II
Day Coach Lighting.
Average horizontal foot-
Installed candles 36 inches
during Description of illuininants above floor
1850 to 1875 Oil lamps — 2 wicks feeding one flame 0.5
— annular wick 1.0
1875 to 1900 Pintsch gas — 4 fish-tail flames 1.5
— 4 mantle cluster 1 .65
1880 to 1900 Carbureted gasoline — center draft 1.3
1900 to date Electric — 50-watt opal dip lamps and flat reflectors 1.4
— 50-watt clear lamps and satin finish
bowl reflectors 3.0
Typical illumination intensities for artificial lighting of certain
classes are indicated in the following table:
Typical Intensities of Artificial Illumination.
Foot-candles
Class As measured through Average Usual range
Street lighting —
Principal streets in cities- -Horizontal plane of
street surface .... 0.4 0.25- 2.0
Important side streets Horizontal plane of
street surface 0.15 0.1 - 0.25
Residence streets Horizontal plane of
street surface .... 0.04 0.01- 0.10
Store lighting Horizontal plane 30
inches above floor 4.0 2.0 - 6.0
Show window lighting Plan of trim 18.0 12.0-25.0
Factory lighting Horizontal plane 30
inches above floor 3.0 2.0 - 6.0
Office lighting Horizontal plane 30
inches above floor 3.0 2.0 - 4.0
Residence lighting Horizontal plane 30
inches above floor 1.5 1.0-3.0
Railway car lighting ... Horizontal plane 30
inches above floor 2.0 1.0 - 3.0
The trend in illumination intensities is upward. Higher ef-
ficiency lamps are being developed, particularly in large illu-
minants suitable for lighting streets, public squares and large
rooms such as factories, armories, etc. With the increase in ef-
ficiency there is coming into our practise a greater insistence upon
good candle-power maintenance, both that inherent in the illumin-
ants and that secured through careful maintenance of the lighting
equipment. Daylight illumination in interiors is perhaps five to
ten times as intense as that provided by our usual artificial light-
THE STATUS OF THE LIGHTING ART 659
ing. As the artificial lighting is improved, attaining greater merit
in respect to diffusion, the tendency appears to be to increase the
intensities. Latest experiments seem to indicate that it is not im-
probable that when we shall have attained more complete knowl-
edge of the principles of good illumination, we shall find that the
intensities now available in the illumination of interiors by day-
light will have to be approached by artificial light in order to
satisfy the requirements. It is apparent further that as appre-
ciation of the beautifying opportunities in illuminating engineer-
ing grows, esthetic considerations impose requirements for greater
light production. Thus the requirements of ocular hygiene and
of esthetics combine in demanding the production of more light
than was formerly required in order to provide acceptable illu-
mination. The growing appreciation of the importance of good
lighting is raising the standard, making a still further general
demand for the production of more light.
DIRECTION OF LIGHT.
Direction of light has not received the same extended con-
sideration as have some other features of the illumination prob-
lem. It is forced upon attention, however, in certain classes of
work where improper direction brings annoying shadow or glare.
In such cases the remedy is sometimes found by changing the
direction of the light.
In general, artificial lighting is provided from ceiling fixtures ill
the center of rooms or bays, though often wall brackets are used
alone or to supplement center lighting. This involves a down-
ward direction of the utilized light. Where the ceiling is em-
ployed as a secondary source, as in indirect lighting, engineering
thought seems to favor designs which will very largely preserve
a downward direction for the light. Daylight illumination of in-
teriors, on the other hand, is very generally from side windows,
and the lighting has a strong component which is almost hori-
zontal, though the direction is usually sjightly downward.
One of the earliest evidences of appreciation of the importance
of proper direction in lighting is the well established tradition
that in reading "the light should come from over the left
shoulder." Many of the tenets of illuminating engineering,
though unrecognized in the formulation of this precept, are evi-
660 TRANSACTIONS I. E. S. — PART II
dent as the underlying cause which has led to its wide dissemin-
ation and general acceptance. It was the outgrowth of reading
experience in a time when a limited amount of reading was done
with the aid of a single illuminant in a room. This homely saying
correctly indicates conditions for reading which avoid shadow
and glare and best contribute to the ocular welfare and comfort
of the reader. Under the conditions of use which obtained when
one or more members of a family read by the aid of light from
an oil lamp, the portability of the light source, and the freedom
of the reader to choose any desired position, made compliance
with the precept entirely practicable.
Modern conditions, involving immobile light sources and fixed
positions for a number of workers in the same room, complicate
the problem severely, and demand much more adroitness for its
successful solution. Indeed were proper direction of light the
only possible solution, as is implied by the ancient precept just
quoted, the problem in many cases would be very difficult.
Recently, question has been raised as to the general propriety
of downward light.* Arguments have been advanced in favor
of a direction of light which is from the side. It is quite possible
that in the near future developments in illuminating practise may
result in less general adherence to a downward direction for
light than characterizes present practise.
Meanwhile, however, the general downward direction of light
is very common. In store and office lighting, reflectors designed
to redirect the light downward are employed very generally.
Only 15 to 25 per cent, of stores and offices employ illuminants
without auxiliaries of some kind. In many of these exceptional
installations the lamps are placed so near the ceilings that a
general downward direction of light is obtained in spite of the
absence of redirecting auxiliaries. Of the 80 per cent, installa-
tions which employ lighting auxiliaries, probably four-fifths, or
about 65 per cent, of all installations employ some device designed
to direct much of the light where it is considered to be most use-
ful.
Under daylight conditions the direction of light is likely to be
from the side, and altogether too little attention is paid to adapt-
* Ives— Some Home Experiments in Illumination— I. E. S. Transactions, June, 1913,
Page 229. -
THE STATUS OF THE LIGHTING ART 66l
ing conditions in order to secure the best illuminating results.
Indeed where a number of persons are at work in a large room
illuminated from the side by daylight, it is well-nigh impossible
to dispose things so that there shall be practical freedom from
shadows and glare.
Direction of light is so intimately associated with diffusion in
contributing to the merit of an installation, that its further con-
sideration may well be included under the following caption.
DIFFUSION OF LIGHT.
Diffusion of light in earlier discussions of illumination prob-
lems was often regarded simply as a quality secured as a result
of an effort to conceal light sources and reduce their intrinsic
brilliancy. Its influence upon the appearance of a room and
of objects in a room was recognized, and its importance from the
ocular standpoint was often considered, though not fully apprecia-
ted. Importance as an element contributing to the reduction of
sharp shadows completed the list of recognized effects of diffusion.
Larger experience in illuminating practise and more recent ex-
periments and research have brought about an increased and
more widespread appreciation of the importance of diffusion as
one means of diminishing glare from reflecting surfaces and
promoting ocular welfare in general. Increased appreciation of
the importance of securing a proper degree of diffusion is perhaps
the most notable development in the knowledge of illuminating
engineering during the past two or three years. A few years ago
the cry was for the use of efficient reflectors which directed the
light downward, increasing the horizontal illumination intensity
for a given light production, or decreasing the amount of light
which would have to be produced in order to provide a given
illumination intensity.
At this time such reflecting devices, while regarded as useful,
are recognized as inadequate unless some means of providing a
fair measure of diffusion is employed. Thus the use of mat re-
flecting surfaces is growing. One year ago, it was found that
among 52 types of simple glass and metal reflectors purchased
upon the open market 38 had been provided with etched or
other diffusing surfaces.
662 TRANSACTIONS I. E. S. — PART II
Where diffusion is lacking, multiplicity of sources is of some
assistance, since in some cases glare may be overcome by in-
creased intensities of light from other directions. Multiplicity
of sources, however, means multiplicity of shadows.
Extremes of diffusion are rarely required. Too much diffu-
sion means characterless illumination and may mean eye-fatigue.
Too little diffusion involves glare. The degree of diffusion which
characterized even the better lighting installations of a few years
ago is now recognized as inadequate. "Semi-indirect" and "in-
direct" lighting succeed in many installations because of the
higher degree of diffusion which their use involves.
Diffusion of light cannot be separated from intensity of light
in discussions of this character. Obviously, light cannot be dif-
fused without loss. Some have felt that with diffused light a
lesser intensity is satisfactory. Tests have been reported which
purport to establish this point. It is very doubtful, however,
if such is the case. With a correct degree of diffusion it is en-
tirely practicable to determine what intensity of light is most sat-
isfactory for a given purpose. With improperly diffused light
it has been found that the same intensity of light is unsatisfactory.
This has been taken to indicate that proper diffusion of light
results in a decreased intensity requirement. When the diffusion
is inadequate, the lighting cannot be satisfactory with any inten-
sity provided by the same source or sources. When subjects
who are being experimented upon are told to increase the intensity
until the illumination is satisfactory in order to compensate for
the lack of diffusion, it is not unnatural that the result should be
a report that increased intensity compensates for the lack of
diffusion, giving rise to the conclusion that increased diffusion
permits decreased intensity. It is submitted that such an effect
has not been demonstrated and there is no reason to suppose
that insufficiency of diffusion can be compensated for in any other
way than by increasing the diffusion. Hence the intensity re-
quirement with properly diffused light cannot be said to be less
than it is with improperly diffused light. Each increase in light
diffusion has brought with it the necessity for increasing the light
production in order to compensate for the loss in efficiency in-
volved in the diffusion.
THE STATUS OF THE LIGHTING ART 663
In one-half to two-thirds of the total installations upon which
reports have been obtained, certain degrees of diffusion have
been obtained, either through the employment of diffusing globes,
by depolishing inner surfaces of reflectors, or by means of in-
direct lighting.
COLOR OF LIGHT.
The modification of light to produce desired colors has been
very generally employed in recent years for decorative purposes.
This usage has been very largely the outgrowth of effort intended
to secure decorative results. The ends to be achieved have been
more or less clearly defined, but the means to be employed have
in general received too little attention. Color in lighting has
been the agent of the artist rather than of the physicist or en-
gineer. In its employment empiric rather than scientific methods
have been followed. A quality of light which must ever com-
mend itself to the artist and which can be employed effectively
only through artistic appreciation, color remains both a physical
and a physiological phenomenon which must be applied scien-
tifically to secure thoroughly effective results.
The art of illumination seems to be upon the verge of a marked
advance due to appreciation of the possibilities of color manipu-
lation in lighting. Illuminants are available ranging in color
from the neon tube through the Moore tube and the metallic
electrode arc to the mecury-vapor lamp among luminescent
light sources. The incandescent light sources offer a narrower,
though material, range of color values. The employ-
ment of these variously colored illuminants in conjunction to pro-
duce color effects is inherently costly and troublesome, and likely
to be practised only in special cases, as in the notable installation
in the Allegheny County Soldiers' Memorial Hall, Pittsburgh.*
The modification of the light of many or all of these illumin-
ants to secure light of desired quality is a field which, now largely
uncultivated, offers opportunities of which the illuminating en-
gineer is beginning to avail himself.
The use of differently colored illuminants for commercial pur-
poses is beginning to be appreciated. In a few dry goods stores,
* I. E. S. Transactions, Vol. VI, 191 1, Bassett Jones, Jr.
8
664 TRANSACTIONS I. E. S. — PART II
small area lighting by illuminants which simulate daylight has
been provided.*
For many purposes we aspire toward light of the color of
daylight. The earliest successful illuminant for this purpose,
namely the Moore carbon dioxid tube, produced light of practi-
cally the desired quality. Later efforts in the modification of light
of electric incandescent lamps, incandescent gas mantle lamps and
intensified carbon arc lamps, have attained some measure of suc-
cess. It is understood that all these forms of artificial daylight
have come into use to a limited extent, though the low efficiency
of all is a deterrent preventing their application upon a large
scale.
We appear to be upon the verge of developments in arc lamps
which are likely to result in relatively high efficiency illuminants
of color values so near to daylight that it would not be surpris-
ing if the slighter modification of light which would be entailed
in order to produce artificial daylight would be accomplished with
a loss which would not reduce the efficiency of the lamps below
the practicable limit, and which, therefore, might result in arti-
ficial daylight illumination upon larger scales. Such a consumma-
tion is to be desired, because artificial daylight will not only be
useful in a practical way, but experience in its utilization is
likely to extend the realm of our practical knowledge of illumin-
ating principles.
Meanwhile it is becoming increasingly apparent that for certain
purposes, as the lighting of residences, ballrooms, etc., white light
is not acceptable under existing conditions. Having become
accustomed for generations to the employment of light in which
the long wave-lengths are more accentuated, we have either
adapted ourselves to that condition, or there is some quality in-
herent in its present application which makes it more satisfactory
for social purposes than white light. The amber mantle as a
substitute for the ordinary mantle employed in gas lighting caters
to this peculiarity and renders the gas mantle lamp far more
acceptable for certain purposes. Among electric incandescent
lamps there is reason to believe that the light of the tungsten fila-
ment lamp might be modified to advantage in order to produce a
* For example see Shalling— Store Lighting, I. E. S. Transactions, January, 1913.
the: status of the lighting art 665
more acceptable color for social lighting. Attention is being given
to these matters, and it is reasonable to state that the art is upon
the verge of developments which within the next few years will ex-
tend employment of color in lighting, and that lighting auxiliaries
will be developed along scientific lines which will yield a wide
variety of tints with a minimum practicable loss in efficiency.
The physiologist, and the psychologist, as well as the artist,
are interested in the color of light. Our knowledge of these
matters is not great, and its possession is not general. While it
appears probable that we are upon the threshold of a general
effort to develop the use of color in illumination along scientific
lines, thereby upholding and encouraging artistic endeavors in
this direction, it remains true that more encouragement should
be offered to the study of the effect of light of different colors upon
the eye and upon the mentality. Only as artist, psychologist,
physiologist and engineer combine, will the employment of color
in illumination be developed to a point of maximum effectiveness.
The most encouraging fact in this connection is the activity of this
Society in promoting the co-operative study of this problem and
in disseminating knowledge accumulated through such study.
STEADINESS OF LIGHT.
Steadiness of light is so important a fundamental that most
modern illuminants employed in indoor lighting are free from
serious objection on this score. Outdoors the require-
ment for steadiness is less rigorous, and illuminants pro-
ducing light which is relatively unsteady find employment more
readily. Among such are, in varying degree, arc lamps and the
flame illuminants, including open flame gas lamps, and some types
of gas mantle lamps. The flame illuminants are inclined to be
unstable when exposed to wind. In modern types of gas
lamps this difficulty is reduced. Among arc lamps, the flaming
arc is particularly unsteady. Improvement in the steadiness in the
light from all arc lamps is effected when the current density at
the electrodes is increased. The tendency to operate these lamps
at higher currents, as noted elsewhere, brings greater stability
in addition to higher efficiency.
Cyclic fluctuations in electric lamps constitute another form
of unsteadiness of light. This effect is the more notice-
666 TRANSACTIONS I. E. S. — PART II
able in the case of arc lamps. Under some conditions it
has been found practicable to operate incandescent lamps upon
25-cycle alternating current circuits. This frequency is about the
lower limit and under some conditions the flicker of lamps so
employed is objectionable. Arc lamp operation upon 25-cycle
current is impracticable where good lighting is the desideratum.
In general, steadiness of light is so obvious a fundamental
that a degree of unsteadiness which becomes obtrusive is rarely
tolerated.
CONTRAST IN ILLUMINATION.
Contrast, here regarded in a comprehensive sense, is a most
important factor in the utilization of light.
Exposed light sources within the field of vision, occasioning
too great contrast with the surroundings to which the eye is
adapted, are detrimental to vision. One of the earliest canons
of illuminating engineering branded exposed light sources as the
cause of ocular discomfort and, under extreme conditions, of
impairment of vision. The development of more powerful and
brilliant light sources has emphasized the importance of this prin-
ciple. Reduction in the specific intensity of light sources was
recognized as a means of mitigation. Tables of specific intensity
of various light sources were prominently featured in the earliest
discussions of illuminating engineering. The low specific inten-
sity, for example, of the Moore tube was urged as one of its
chief merits. But that exposure of light sources was not gen-
erally recognized as a serious menace to eyesight and a source of
discomfort, is attested by the designs of lighting fixtures and
glassware of the period. Catalogues of lighting equipment fur-
nished by the leading manufacturers at that time show designs
which in the light of our present practise appear almost barbarous
in their defiance of this first principle of illuminating engineering.
The evil effects of exposed light sources include, as previously
mentioned, ocular discomfort and, under certain conditions, im-
pairment of vision in the sense both of injury to eyesight and
diminished visual power. These effects in rather loose termin-
ology have been attributed to glare. A number of causes and
effects differing materially in character have been classed as
glare without proper distinction, due to their nature. The study
the; status of the; lighting art 667
of glare and its evil effects, insofar as these are concerned with
the exposure of light sources, received a notable impetus in the
early years of this Society's history. This was particularly true
of glare as an agent which diminishes visual power.* Following
upon the discussion of glare in the sense here considered, there
came a more general appreciation of the advantages of concealing
light sources. Leading manufacturers of lighting appliances dis-
continued some of the most prominent offending designs and the
newer designs evidenced more attention to this important aspect
of the problem. While on all sides one finds evidence of continued
neglect of this important fundamental, yet we may feel that, at
least by lighting practitioners, there is a thorough understanding
of the serious nature of exposure of light sources, and that a real
effort is being made to eliminate it from our practise. It is now
generally recognized that entire concealment of the light source
as in indirect lighting, reduction of its brilliancy by the aid of
diffusing media, as in much of the direct and "semi-indirect"
lighting, or the removal of the light source from the ordinary
field of view, is an essential to good lighting. Very generally
these principles are being put into practise. Flat reflectors, which
exposed the light source, are being replaced by bowl reflectors
which conceal it from ordinary view ; manufacturers of gas
lamps report growing use of diffusing rather than clear globes;
manufacturers of electric lamps report increasing use of frosted
lamps ; the use of indirect and "semi-indirect" lighting fixtures is
growing rapidly ; new installations of street lamps for civic light-
ing are very generally mounted higher than was the practise a
few years ago; the proportion of bare lamp installations is de-
creasing.
Exposed light sources of the more usual types are more bril-
liant than is the sky, and surroundings are apt to be less bright
under artificial light than they are in the daytime. Thus greater
contrasts prevail in artificial lighting. In the better lighted instal-
lations, however, improvement is now being effected both through
the reduction of the specific intensity of the sources and through
the increase in the brightness of the surroundings.
Glare from reflecting surfaces due to specular reflection is
occupying much thought of investigators and practitioners at the
* A. J. Sweet, Journal Franklin Institute, May, 1910.
668 TRANSACTIONS I. E. S. — PART II
present time. This again involves the question of contrast.
Indeed, all manifestations of glare are but little more than im-
proper contrast. A printed page of glossy paper may give rise
to a serious condition of glare due to specular reflection. This
means that one views the imperfectly reflected image of a light
source which, in contrast with the surroundings, is of excessive
brightness. The result is ocular discomfort. If glossy ink has
been used in printing, it may be difficult or even im-
possible to discern the imprint. This is a case in which the
excessive specular reflection which causes discomfort by too great
contrast with the surroundings is complicated by too little local
contrast to permit of reading the glossy black letters, which reflect
specularly almost as well as the surrounding surface of the paper.
This element of contrast is receiving a fair share of the atten-
tion which it merits, and our knowledge of the principles involved
is growing rapidly. This is followed by improvement in the
design of new installations in which such knowledge is applied.
In general, therefore, it may be said that improper contrast which
manifests itself as glare either due to exposed light sources or to
reflections from specular surfaces is rapidly being brought under
control in the installations which are receiving illuminating engi-
neering attention. The lesser degrees of difficulties involved in
improper contrast are, however, not so thoroughly understood or
appreciated; our knowledge appears to be in need of con-
siderable extension and our practise appears to suffer from lack
of proper care for this important feature of the utilization of
light.
CONGEUITY.
The past 2 or 3 years have witnessed a marked increase in gen-
eral appreciation of the importance of the esthetics of illumina-
tion. It is becoming generally recognized by lighting practitioners
that lighting equipment as well as the illumination produced must
be in harmony with the character of the premises which are
illuminated.
In the early development of the science and art of illumination
it was but natural that the physical and engineering aspects should
receive first attention, for these underlie the entire art. It was
THE STATUS 0E THE LIGHTING ART I 69
but natural that first developments should evidence incongruities.
This period of illuminating engineering is now being outgrown.
The art is now approaching the period of adolescence in which
the acquirement of knowledge is rapid and preparation for mature
effort is the keynote. At this time it is being appreciated that
congruity in the utilization of light in any installation is a desid-
eratum of first importance. It is just as inartistic to locate highly
embellished ornate and inefficient fixtures and glassware in a
machine shop, as to hang tin shades in an elegantly furnished
drawing room. Error, however, occurs commonly in the use of
inartistic lighting equipment where artistic design is required,
and it is in this aspect of general practise that one of the great
needs for improvement is apparent.
It is recognized that the illuminating engineer should be con-
versant with and appreciative of art in its several phases if his
work is to reveal in an intelligent sympathetic manner the best
of the design. Some instances are encountered where only the
artist can do justice to the requirements. In such cases the
illuminating engineer should be guided by the artist, and he should
be able to accomplish the desired end more effectively and with a
lower cost than the artist can. The latter fully appreciates what
he desires, but presumably has not familiarized himself with
lighting technique to the extent which enables him to accomplish
his purpose so well as can the illuminating engineer. It is believed
that the events of the past 2 or 3 years have contributed toward a
somewhat better mutual understanding and that in the very near
future we shall enjoy a fuller measure of the needed hearty
co-operation of architects, decorators and fixture designers,
thereby achieving improvements in the artistic phases of illumina-
tion which are so much needed at the present time.
In this connection it may be noted that a most important factor
in the betterment of illumination conditions, artistically, is im-
provement in the design of stock fixtures and stock lighting
auxiliaries. Any given type of fixture must of course be designed
for one set of conditions and cannot be expected to be congruous
in a wide variety of installations. It may, however, be made
tasteful in itself, free from objectionable features and well
adapted for general use under average conditions of the class for
which it is designed. When considering the artistic aspects of
670 TRANSACTIONS I. E. S. PART II
illumination, it is not well to restrict thought to individual designs
which of course are expected to surpass stock fixtures being at
once more pleasing and congruous when applied in the installa-
tions for which they are designed. It should be remembered that
a 5 per cent, improvement in stock fixtures affecting large num-
bers of people may be of much greater value than a 100 per cent,
improvement in a distinctively designed installation.
Residences and certain other classes of installations may be
said to be equipped very generally with fixtures which cannot be
approved either from the artistic, hygienic or efficiency stand-
points. This is particularly true of installations which date back
5 years or more. It is especially true of combination fixtures
designed to permit the use of either electric or gas illuminants.
The presence of these fixtures which are neither tasteful in them-
selves nor consistent with the standard of good taste exhibited in
the furnishing of rooms in the middle and better class residences,
constitutes a barrier to the improvement of lighting conditions in
residences. The first step toward a general improvement in such
lighting conditions would appear to be the wholesale replace-
ment of such fixtures. Such is the crying need of the time. The
way of accomplishing this is not clearly indicated. The extension
of lighting improvements to residence and to some other classes
of lighting is dependent upon the displacement of such fixtures.
It is fortunate that displacement of incongruous lighting equip-
ment by equipment which would be suitable from the artistic or
efficiency standpoint would bring about automatically an im-
provement in installations from the standpoint of ocular welfare.
If the requirements for artistic and hygienic equipment were
diverse or incompatible, the situation would indeed be difficult.
As it is, the happy concordance of hygienic and artistic require-
ments constitutes a demand for betterment which cannot be
ignored.
HYGIENE AND SAFETY.
In ordinary lighting practise the qualities of illuminants are
such that laws of hygiene are rarely transgressed. Much has
been written regarding unsanitary effects of various illuminants.
but usually it has been found that such articles are the work of
the press agent of some manufacturer or promoter of rival- il-
THE STATUS OF THE) LIGHTING ART 67 1
luminants. Ordinarily, ventilation is sufficiently good to avoid
deleterious effects from noxious gases. Most artificial illuminants
are not sufficiently strong in ultra-violet light to be injurious.
The quartz mercury-vapor lamp perhaps forms an exception,
but the heavy glass globe which is invariably used with it when
employed for lighting purposes is sufficient protection.
On the other hand, the use of light promotes sanitation in
securing greater cleanliness wherever it is applied. Likewise the
more liberal and judicious use of light promotes safety. This is
being recognized in the industries among the larger and more
progressive corporations, and more special attention is being
devoted to promoting safety by the use of light.
Of much greater importance than general hygiene is ocular
hygiene in its relation to illumination. It is not from some delet-
erious quality of the light of a particular kind of illuminant that
harm to the eyes results, but rather from the misuse of light, irre-
spective of the illuminant. The efforts which are now being
made to promote the correct use of light, particularly in the
home and in schools, are of incalculable advantage to the public
in safeguarding especially the eyes of children during the period
of immaturity when they are more susceptible to the ill-effects
of misuse of light.
Ocular hygiene is being investigated in a number of labora-
tories. Generally, conditions of visibility are judged by means of
a determination of the threshold visibility value. This involves
the determination of either minimum light intensity for visibility,
minimum size of object viewed, minimum contrast which can be
perceived, or of the time element in the perception of objects at
the threshold value. While it perhaps remains to be determined
how far the results of such investigations may be considered
applicable to practical lighting conditions, yet it is undoubtedly
a fact that the information which is being made available as the
result of such investigations is advancing the science of ocular
hygiene and contributing largely to knowledge of the principles
of good lighting. For the further promotion of ocular welfare
there is need of further research in which the combined efforts
of illuminating engineers, ophthalmologists and psychologists
should govern the nature of the investigation. Also there is need
6/2 TRANSACTIONS I. E. S. PART II
of the application of the results of such investigation to practical
lighting conditions. Conditions appear to be such as to warrant
the assumption that the next few years will witness considerable
advances along both these lines.
COSTS.
It is usually considered that the cost of artificial lighting
includes —
Cost of lighting equipment.
Cost of maintaining equipment including interest, depre-
ciation, etc.
Cost of fuel or energy required for operating the system.
Cost accounting in illumination work is well handled in larger
organizations and possible efficiency improvements are therefore
considered intelligently. Much lighting, however, is not organ-
ized and the expense is borne by those who do not handle accounts
intelligently and who are not in a position to judge of the ulti-
mate efficiency which may be obtained in lighting by the employ-
ment of the several available systems. Thus the use of open
flame gas burners and of carbon and Gem lamps continues in
many installations where the more efficient mantle burners and
Mazda lamps should be used.
Artificial lighting is very inefficient, due first, to the low effi-
ciency of energy transforming devices, and second, to the low
light production efficiency of illuminants. To this in many cases
must be added unintelligent utilization of light. In spite of its
low efficiency, however, the cost of artificial lighting is small.
The following lighting costs are suggestive in this connection :
Approximate cost of artificial lighting
Class of lighting installation in proportion to total operating costs
Small wage earner's home — ratio of cost
of lighting to total income I per cent.
Well conducted large manufacturing es-
tablishments which are well lighted
with modern illuminants — ratio of cost of
artificial lighting to total cost of output
exclusive of selling expenses J 3 to *4 of 1 per cent.
Large retail mercantile establishments —
ratio of total lighting cost to total sales Probably less than i.oper cent.
Small stores — ratio of total lighting cost to
total sales 2 per cent.
Modern loft buildings 1 to 2 per cent.
THE STATUS OF THE LIGHTING ART 673
Street lighting appropriations by municipalities are of the order
of 60 cents to $1.00 per inhabitant per year.
In comparison with the benefits conferred by artificial lighting
both in the way of added commercial advantage and extended
opportunities in education, social life and recreation, the cost of
artificial lighting is remarkably low. In many cases where its
beautifying influence is of paramount importance, the cost is
immaterial.
In this country there is a strong sentiment in favor of corporate
work in the promotion of employees' welfare. The tendency
of the American business man toward organization and scientific
management was never stronger than at the present time. From
both viewpoints, artificial lighting is an important factor and in
both respects its cost is small as compared with the advantages
derived.
The cost of artificial lighting has been rapidly reduced as a
result of the development of improved illuminants and of the
greater economies which have been effected in manufacture and
operation. Doubtless it will be still further reduced in the near
future. This reduction in cost has been largely automatic and the
outgrowth of economic conditions.
With the reduction in the cost of lighting has come betterment
in lighting conditions. Much that is advanced in the lighting
practise in this country is due to the progressive enterprising
manner in which manufacturers of illuminants and lighting com-
panies have conducted their businesses. In the future we must
look to such organizations to bring about improvements in light-
ing practise which are needed if the work of this Society is to
be applied for public benefit.
Only to an extent so small that it is negligible has the decrease
in the cost of lighting been effected through legislation. Recently,
however, in a few cities recourse has been had to rate legislation
affecting public service corporations .in consequence of which rela-
tively low rates have been imposed. These low rates imposed
by city regulation, actuated often by political rather than by
economic motives, constitute a menace to the success of this
Society's work insofar as it may affect the general public. Only
as corporations engaged in supplying lighting service earn a
674 TRANSACTIONS I. E. S. — PART II
fair return upon their investments can they be expected to con-
tinue operation along broadly progressive lines. Only as such
policies shall prevail among manufacturers of lighting appliances
and in the management of lighting companies, may we hope for
rapid progress in the further improvement of lighting conditions.
Commercial organizations have great potentialities for good, and
we can rely upon a continuance of their assistance only if the
further reductions in the cost of lighting are effected in accord-
ance with the development of the economic situation rather than
in accordance with the artificial conditions of political regulation.
The foregoing brief review is based in part upon information
which has been made available by replies to the questions which
were issued as a part of the lighting survey. These replies are
extensive and numerous, affording in the aggregate a large amount
of statistical information. They apply, however, to so small a
percentage of the total lighting industry as to make it seem unwise
to present a complete summary of the facts which they make
available. All the information which has been obtained in this
way is available to the society and may be very useful in the
furtherance of statistical work along this line.
In the preparation of the detailed information which has been
made available as the result of this lighting survey, the writer
has been favored by the kind co-operation of individuals and
companies so numerous as to make it impracticable to list them
in this connection ; it is possible only to express deep apprecia-
tion. The loyal disinterested assistance of Mr. N. D. Macdonald
of the Electrical Testing Laboratories should be recorded
especially.
CORRELATED MATTERS.
In addition to a review of knowledge and practise in respect
to the several qualities of light and features of utilization, no dis-
cussion of the status of the lighting art would be complete with-
out reference to a number of supplementary factors which have
been influential in establishing the present status and which may
be looked to for assistance in advancing that status. Prominent
among these are educational agencies, commercial organizations
in the lighting field, the Illuminating Engineering Society, the
attitude of related professions and industries, and photometry.
THE STATUS OF THE LIGHTING ART 675
EDUCATIONAL AGENCIES.
In the way of technical education, universities and colleges
are devoting more attention than formerly to the principles of
illumination and to photometry. The Johns Hopkins University-
I. E. S. lecture course on illumination did much to arouse ped-
agogic interest in this subject, and the Society's Committee on
Illuminating Engineering Education is engaged in an effort to
apply the lecture course and promote the further specialization
along this line in university education.
It is of prime importance that those who are in a position to
influence the design of lighting installations should possess a
knowledge of the principles of good illumination. Educational
work in this connection is being done on a limited scale by some
of the larger manufacturers of lamps and lighting appliances
through lectures, the publication of bulletins, and through free
consulting engineering advice to customers. A number of the
larger lighting companies contribute like service to their cus-
tomers. The Commercial Section of the National Electric Light
Association and the National Commercial Gas Association have
done commendable work in the publication of bulletins on cer-
tain phases of lighting work. This Society has in course of pre-
paration lectures on several classes of lighting the manuscript
of which, accompanied by lantern slide illustrations, is to be made
available for presentation wherever required.
Supplementing these efforts is the dissemination of elementary
knowledge of good illumination direct to the public through the
illumination primer prepared by this Society. Through the
efforts of the Society and the co-operation of other organizations
this has been issued throughout the country in quantities which
aggregate about 250,000.
Knowledge is power. If the simple elementary truths in re-
gard to lighting could be imparted to the public, lighting practise
the country over would be advanced tremenduously. Through
the several agencies just mentioned it is believed that great
strides are being made in this direction.
COMMERCIAL ORGANIZATIONS IN THE LIGHTING FIELD.
In this country the very general enterprise and progressiveness
of the leading manufacturers of lamps and lighting appliances is
676 TRANSACTIONS I. E. S. — PART II
an asset upon which we are to be congratulated. A number of
such organizations maintain laboratories for research work as
well as engineering and educational departments. Thus these
manufacturing organizations contribute to the advance of the
lighting art in three ways. First, by developments made pos-
sible through research and invention; second, by the incorpora-
tion in the design of their product of the information made avail-
able through research in their own laboratories and elsewhere;
third, by the general dissemination of the facts which govern
good practise.
The progressive attitude of lighting companies, particularly
those in the larger cities is likewise commendable and is an im-
portant factor in lighting improvement. Through the training
of solicitors in lighting fundamentals and the maintenance of il-
luminating engineering departments, as well as through adher-
ence to the free lamp renewal policy, and the promotion of the
use of the most efficient lamps, the contributions of such or-
ganizations is large and important.
Another agency not to be ignored is the papers and discussions
on illumination which are now included very generally in the pro-
grams of electric and gas associations, and to some extent in the
programs of professional bodies. Still another agency is the
technical journals, which now very generally feature illumination
discussions, thereby doing much to promote interest and advance
knowledge in illumination affairs.
THE ILLUMINATING ENGINEERING SOCIETY.
In his inaugural address the writer endeavored to review the
growth of the society and to state its functions as these are
now recognized. At first, scientific discussions predominated be-
cause knowledge must precede application, and knowledge of il-
lumination principles was meager. The tendency during recent
years has been to supplement the scientific work with practical
discussions of lighting practise and with efforts along educational
lines.
The only organization in the country which deals exclusively
with light, this Society now commands a fair measure of respect
from the older national technical societies. It is beginning to
realize returns upon its large expenditure of time and effort in
THE STATUS OF THE LIGHTING ART 677
co-operative work with other technical societies. It is undoubt-
edly one of the principal influences which make for improvement
in the lighting art.
THE ATTITUDE OF RELATED PROFESSIONS AND INDUSTRIES.
That the efforts of the Illuminating Engineering Society to
improve lighting practise are not lacking in support by other or-
ganizations has already been made apparent. It has been indicated
that large and progressive corporations are devoting attention to
the securing of good illumination as a part of good management
and also in connection with welfare work for employees. Light-
ing companies, both gas and electric, are showing revived appre-
ciation of the importance of the lighting aspect of their business.
While in both industries the lighting load is decreasing in pro-
portion to the total load, yet it is recognized that through its
lighting service the company comes into contact with the greatest
numbers of its public, and excellence of service in this connection
is its best possible advertisement.
As an illustration of the present view of lighting companies, it
may be said that in the 1913 report of the Lamp Committee of
the Association of Edison Illuminating Companies, considerable
space was devoted to the importance of promoting good illumina-
tion as a means of cultivating good public opinion and conserving
revenues.
Perhaps the most interesting and promising features of the
replies received in connection with the lighting survey which is
described herein is the very evident interest manifested by oph-
thalmologists throughout the country in the subject of illumina-
tion. There is undoubtedly reason to believe that the society will
enjoy a larger measure of co-operation from ophthalmologists in
the immediate future than has been accorded in the past, and
this promises well for improvement in home conditions of light-
ing.
It must be said that the co-operation of architects in the devel-
opment of lighting practise is in 'need of further cultivation.
Attempts during the past year to promote further co-operative ef-
fort with organizations of architects have come to naught, although
encouragement was derived from the cordial attitude met in all
cases. It is hoped that a start may be made during the coming
678 TRANSACTIONS I. E. S. — PART II
year which will result in securing recognition of mutuality of
interest and a greater measure of that co-operative effort which
is so necessary to the improvement in lighting practise in build-
ings in which the architect requires consistent lighting treatment.
PHOTOMETRY.
To the development of a science, proper measurement of the
quantities involved is essential. In the growth of knowledge of
illuminating engineering, photometry has played an important
part. At the present time the status of photometry is definitely
established, and very generally recognized. In the manufacture
of illuminants photometry is resorted to in rating and efficiency
adjustments. In taking the candle-power of gas it finds one of
its most general applications. In spite of the increasing strength
of the movement to rate gas upon calorific rather than photo-
metric value, the candle-power basis still obtains very generally.
In the production of incandescent electric lamps, photometry
has found one of its widest applications. The standardization of
manufacturing methods, however, is resulting in the abandonment
of photometry for the rating of individual Mazda lamps, though
•the photometry of samples from each batch manufactured still
continues to be a regular part of the manufacturing procedure.
In the manufacture of other illuminants, photometry is a regular
part of the engineering work. In the study of illumination, pho-
tometry is practised very generally by illuminating engineers.
Accuracies obtained in practise in the various classes of work
are of the order indicated in the following table:
Typical Accuracies Obtaining in Photometry of Light Sources
Which Involves no Large Color Differences and no
Serious Variations in Light Intensity.
Per cent.
In taking the candle-power of gas 2
In routine rating of carbon and Gem filament lamps ±4
In routine commercial testing of incandescent electric
lamps and incandescent gas lamps — ±2
In precision photometry ±}4
In heterochromatic photometry lower orders of accuracy are
encountered, depending upon the extent of the color difference
and the methods employed in determinations of intensities.
Where material variations in intensities are involved, further in-
THE STATUS OF THE LIGHTING ART 679
accuracies may result due to personal pecularities of observers
in recording observations under such conditions. Here again
the accuracy which is obtained depends largely upon the extent
of the unsteadiness of the light source and the methods of test
employed.
A matter of immediate importance to photometricians is the
adoption of a standard for use in heterochromatic photometry.
In view of the great variety of light sources which are available,
and their differing color characteristics, it is important that stand-
ards be adopted for working purposes, even though such stand-
ards cannot be relied upon for ultimate accuracy. In the adop-
tion of such standards it is believed that the entire laboratory
resources of the country should be utilized. This is one of the
next important steps to be taken in photometry.
conclusion.
The foregoing is a very inadequate review of the condition in
the field which this Society seeks to cultivate. Circumstances
beyond control interfered with the writer's intention to make this
survey as comprehensive and accurate as possible. Instead of
presenting a complete and satisfactory statement which might
be placed upon file as a record of the status of the lighting art,
it becomes necessary to offer this review as a first step toward
the preparation of such a record. However, discussion of the
subject with members of this Society, and experience in prepar-
ing this review, have brought conviction that a continuation of
this effort to compile an adequate compendium of the knowledge
of the art and of prevailing practise would result in much good,
and would repay the large effort which would be required. It is
therefore without apology for the obvious inadequacy of this
presentation, but rather with the expressed hope that others may
consider the advisability of undertaking a more thorough and
authoritative treatment that this survey of lighting conditions is
presented.
Our knowledge of the principles of good illumination, though
lacking in many essentials, is still considerable. The advances
made since the organization of this Society in 1906 are gratify-
ing. The condition for further development of the underlying
principles is most hopeful, in view of the investigation and
9
68o TRANSACTIONS I. E. S. — PART II
research which are being carried on by a goodly number of our
members. Some of this research in particular, is organized upon
such a plan as to warrant a most optimistic view of the probable
extension of knowledge in this field in the next few years.
The standard product of manufacturers is improved rapidly
in accordance with additions to knowledge of lighting principles.
Such improvements in the case of lamps or appliances which have
limited life and must be replaced periodically, find their way grad-
ually into most installations. In the case of appliances which do
not have to be replaced periodically, such as fixtures, lighting
glassware, etc., the improvements are not applied so generally in
lighting practise. It is perhaps a misfortune that such appli-
ances do not have a limited life. If fixtures, reflecting glassware,
etc., would automatically disintegrate after a reasonable period
of service, the commercial incentive offered to manufacturers to
improve lighting equipments would be even greater than it is
to-day, and the public, would benefit more generally as a result
of the improvement which is effected in more recent designs. As
it is, new installations benefit, but existing installations in large
part obtain little or no advantage from the more recent develop-
ments.
Perhaps the biggest problem to be solved by the Society in its
effort to secure the general improvement of illumination in all
classes of installations is that of displacing antiquated lighting
appliances. It would appear that we must look to those who
have possible commercial advantage to derive for the display of
enterprise which is essential to the general application of better
knowledge of illumination principles to general lighting practise.
These are the manufacturers of the equipments, and the lighting
companies who supply the service. Co-operation between these
two classes alone can bring the rapid improvement in the lighting
art which conditions demand and which it is the avowed object
of this Society to promote. Improvement in artificial lighting
which involves a substantial increase in its cost, does not weigh
as heavily in the scale of expenditures as would a corresponding
increase in many other items of expense.
Given a condition in which it is apparent that the public and a
number of important commercial interests will benefit alike by
improvement in lighting practise, and a total cost of lighting
THE STATUS OF THE LIGHTING ART 68l
which is insignificant in comparison with the benefits conferred
by the lighting, and a cost of improvement in such lighting which
is small in comparison with the advantages realized through such
improvements, it would appear that there is no insurmountable
obstacle in the way of attaining the desired result of generally
improved lighting. All that is necessary is a co-ordinating in-
fluence and the necessary conviction as to the results to be ob-
tained. Herein lies this Society's opportunity.
There is reason to be dissatisfied with progress. Practise lags
inexcusably behind knowledge and ideals. It will be conceded
that good illumination is greatly to be desired, the same conclu-
sion being reached whether the viewpoint is dominantly commer-
cial, esthetic or humanitarian. In improved lighting, benefits
accrue to the public, to consulting engineers, to lighting com-
panies, to manufacturers of lamps, to manufacturers of auxil-
iaries, and to manufacturers of fixtures. To improve illumina-
tion conditions, the public must be aroused to an appreciation of
the advantages ever associated with such improvement. This can
be done if all the interests just mentioned can combine and co-
operate to support the propaganda of this Society. The enthu-
siasm and conviction which would necessarily characterize such
a movement would constitute an irresistible force which would
be certain to accomplish the purpose of awakening public interest
and bringing conviction of the benefits which would be sure to
follow. How are we to enlist the support and co-operation of
the commercial interests whose participation in such a forward
movement is essential to its success? This is the practical ques-
tion which this Society must answer. Already the beginning has
been made, and the Society now enjoys a measure of sympathy
and support from the commercial interests with which it has
never before been favored. However, it is necessary to con-
vince the lighting industry of the potentialities of the situation
before any really effective campaign can be waged on a large
scale. What measures can be adopted for bringing this about ?
Two are suggested herewith.
First, it is recommended that the Society supplement its splen-
did work of research, discussion and education, by devoting
special considerations to ways and means of adopting its work to
682 TRANSACTIONS I. E. S. PART II
meet the requirements of the commercial interests to which we
must look for extension of improved lighting practise. A com-
mittee on commercial application would doubtless do much to
further this cause.
Second, the members of this Society individually may accom-
plish results which in the aggregate will be far-reaching in their
effect in popularizing good illumination. Lighting which is in-
adequate, inartistic, unhygienic or inefficient is often observed in
the homes of the members of this Society. Comment brings
some such reply as that "He that makes shoes goes barefoot him-
self." It is submitted that this state of affairs is a reproach under
which no member of this Society should remain. It behooves
each of us to combine practise with precept. Surely there is no
member of this Society who does not feel assured that by light-
ing his home properly, he will contribute to the happiness and
welfare of the members of his family, perhaps more largely than
is possible with a similar expenditure in any other way.
Then what excuse can there be for the member of this Society,
who presumably is informed in these matters, who recognizes the
importance and the possibilities of good lighting, and yet
permits the retention in his home or office of improper lighting
equipments ?
Furthermore, it may be assumed that most of the members of
this Society have more or less commercial interest in the ex-
tension of good lighting. On narrower and more sordid grounds,
is it not incumbent upon such members to promote good lighting
practise by example ?
Should we not attend to the installation in our homes of mod-
ern lighting equipment which is in reasonable conformity with
the latest tenets of illuminating engineering? Also should we
not employ such equipment freely, to the end that in our homes
we may at all proper times exemplify the faith which is within
us? How great an influence upon the community would be ex-
erted if every man who is interested in the furtherance of our
propaganda would so illuminate his home that it would demon-
strate to all who observe, the advantages of good lighting! It
ought to be possible to say of the members of this Society — "By
their lighting ye shall know them."
ANNUAL REPORT OF THE GENERAL SECRETARY 683
ANNUAL REPORT OF THE GENERAL SECRETARY
FOR THE FISCAL YEAR ENDING SEPTEMBER 30, 1913.
The past fiscal year of the Society has been one of general
expansion. The educational and co-operative work started in
the previous year has been continued and extended, while new
lines for developing the Society and enlarging its sphere and in-
fluence have been followed with gratifying results. Perhaps in
no other year has such a general concerted effort been made to
place the Society on a broad and firm basis for future service.
Some of the results achieved are briefly recounted in the follow-
ing report.
1. FINANCES.
The financial status (see auditor's report on another page of
this issue) of the Society has been improved somewhat by the
acquisition of sustaining members. Although the expenses have
increased because the work of the Society has been conducted
along more extensive and useful lines, the revenue has been
slightly in excess of the expenses.
Under the following captions, Income and Expense statistics
are given to show the extent of the various sources of income
and expense. The per cent, figures which are fairly indicative
may be used for comparisons with other years; but the amounts
are not to be compared with similar figures without considering
the fact that the past fiscal year was of only nine months dura-
tion.
Income. — About 63 per cent, of the total income was derived
from members' dues, and 8 per cent, from sustaining members.
Both sources netted $5,897.08, or 71 per cent, of the total rev-
enue. Only about 28 per cent, or $2,359.41, therefore, was ob-
tained from other sources, including $1,097.14 or approximately
13 per cent, from advertising.
Expenses. — An analysis of the expenses shows that $3,768.84
or 46 per cent, was expended for general office salaries, rent,
supplies, etc. The next largest item was that of the Transac-
tions, $1,844.72, or 22 per cent, of the total expenses.
The latter sum is relatively 25 to 30 per cent, higher than the
cost of the Transactions for a similar period of last year, on
684 TRANSACTIONS I. E. S. PART II
account of the new form and style which was adopted at the
beginning of the present year.
Exclusive of the latter two items, $2,562.09, or 31 per cent,
was required for general Society expenses.
Another interesting fact is that the total revenue from the
dues of members and sustaining members, amounted to 72 per
cent, of the total expenses. Without the revenue from sustain-
ing members, the income from dues was only 63 per cent, of the
total expenses.
The surplus of $1,453.94 as of January 1, 1913, has been in-
creased to $1,865.40 as of September 30, 191 3.
The total assets of the Society are shown to be $6,838.18,
against which there are liabilities of $4,972.78.
II. MEMBERSHIP.
A gain was also made in the membership. One hundred and
forty-eight applications and re-instatements were received ; while
the defections totaled 86 — a net gain of 62 members.
In addition to the foregoing increase 21 sustaining members,
whose names are listed below, were elected. This class of mem-
bership was created by a change in the constitution which became
effective at the beginning of the present fiscal year. It is gratify-
ing to note in passing the number of organizations which have
so promptly applied for membership.
Name Date of election
Electrical Testing Laboratories 3/T4/T3
Holophane Works of General Electric Company 3/14/13
The Edison Electric Illuminating Company of Boston • . 3/14/13
The New York Edison Company 3/14/13
The Philadelphia Electric Company 3/14/13
The Edison Electric Illuminating Company of Brooklyn 4/11/13
Commonwealth Edison Company 4/11/13
Macbeth- Evans Glass Company 4/1 1/13
Westinghouse Lamp Company 4/11/13
Boston Consolidated Gas Company 4/1 1/13
National Electric Lamp Association 4/1 1/13
Benjamin Electric Manufacturing Company 4/11/13
United Electric Light & Power Company 5/ 9/13
Welsbach Company 6/13/13
Consolidated Gas, Electric Light and Power Company
of Baltimore 6/13/13
Alexalite Company 9i22>il$
ANNUAL REPORT OF THE GENERAL SECRETARY 685
Name Date of election
Cooper-Hewitt Electric Company 9/23/13
Jefferson Glass Company 9 23/13
Little Rock Railway & Electric Company 923/13
Pittsburgh Lamp, Brass & Glass Company 9/23/13
The Leeds and Northrup Company 9/23/13
The sources of membership gains and losses are shown in
the tabulation given below :
Sections Un-
affiliated
Phila- New Pitts- New and
delphia York Chicago burgh England foreign
Total number of members
at beginning of year,
January 1, 1913 338 399 209 159 95 135
Total applications received 13 59 33 14 2 27
Defections (resignations
and deceased members) 9 29 15 13 n 9
Total number at end of
year 342 429 227 160 86 153
It will be observed that the only loss of membership was in
the New England Section. From territories without the juris-
diction of sections, 27 applications and 9 defections have been
received, making a net gain of 18 members.
Through death the following members were lost:
Carrigan, Howard F.
1538 First National Bank Building, Chicago, Ills.
Douglass, David
Eau Claire Gas Light Company, Eau Claire, Wis.
Schniewind, Dr. F.,
6 Church Street, New York City.
A geographical distribution of the Society's members is shown
in the following tabulation :
Illuminating Engineering Society Members.
United Stales.
Sustaining
Members members
Alabama 2
Arkansas 1
California 19
Colorado 7
Connecticut 13
District of Columbia 17
Florida 3
686
TRANSACTIONS I. E. S. — PART II
United States — {continued).
Members
Georgia 7
Illinois 129
Indiana 20
Iowa 10
Kansas 6
Kentucky 2
Louisiana 4
Maine 4
Maryland 21
Massachusetts 76
Michigan • 21
Minnesota 9
Missouri 17
Nebraska 4
New Hampshire 4
New Jersey 91
New York 339
North Carolina 2
Ohio 92
Oklahoma 2
Oregon 7
Pennsylvania 34S
Rhode Island ... 5
South Carolina 2
South Dakota 2
Tennessee 1
Texas 4
Utah 4
Vermont 1
Virginia 6
Washington 5
West Virginia 3
Wisconsin 26
i,335
Members not in the United States.
Canada 21
England 17
France 3
Germany 6
Hawaii 1
Japan 1
Mexico 1
Panama • • 1
Philippines •" 2
South America 8
Spain 1
62
Total Members 1.335 + 62 = 1,397
Sustaining
members
ANNUAL REPORT OF THE GENERAL SECRETARY 687
III. SECTIONS.
The following table of comparisons gives a bird's eye view
of the work and progress of the several sections of the Society :
Phila- New Pitts- New
delphia York Chicago burgh England
Total members Sept. 30,
1913 342 429 227 160 86
Net changes during year +4 +3° + *8 +1 — 9
Number of meetings held 65564
Number of papers printed
in Transactions 2* 5 1* 3* 3
Average attendance at
meetings 121 144 68 37 35
Total expenses, 9 months $130.88 $245.03 $217.04 $149.73 $53-78
* One paper pending publication.
Each one of the several sections has had one or more joint
meetings with other Societies. These meetings have for the most
part been devoted to popular and elementary discussions of
lighting subjects. Consequently the contributions to the Trans-
actions have been small compared with previous years. These
meetings, however, which were intended to be of an educational
character, have strengthened the prestige of the Society and won a
great deal of respect for the science and art of illumination from
many societies which have hitherto had only a vague idea of
what is illuminating engineering. In general the success and re-
sults of section activities have been very gratifying.
LOCAL SECRETARIES.
To extend the work and influence of the Society to those cities
which do not have sections, several local secretaries have been
appointed. The local secretaries are :
State City Name and Address
California Los Angeles R. H. Manahan, City Electrician
California San Francisco. .'. . F. Emerson Hoar, Railroad
Commission of State of Cali-
fornia, S32 Market Street.
Colorado Denver G. E.Williamson, Denver Gas &
Electric Light Co.
Georgia Atlanta William Rawson Collier, Georgia
Railway and Light Co.
688 TRANSACTIONS I. E. S. PART II
State City Name and Address
Minnesota Minneapolis G. D. Shepardson, University of
Minnesota, Minneapolis,
Minn.
Minnesota St. Paul A. L. Abbott, Northwestern
Electric Equipment Co.
Washington Seattle Fred. A. Osborn, University of
Washington.
It is expected that these representatives will endeavor to pro-
mote occasional meetings under the joint auspices of the Illumi-
nating Engineering Society and local organizations with a view
to fostering interest in lighting matters. Such activities should
lead eventually to the organization of sections in the given cities
or territories.
TRANSACTIONS.
Eighteen papers on various phases of illuminating engineering
have been published in the Transactions during the past year.
From the list of titles given below it will be noted that these
papers include a wide range of subjects. Taken together they are
fairly indicative of the field covered by the science and art of
illumination.
Department Store Lighting.
Influence of Colored Surroundings on the Color of the Useful Light.
The Theory of Mercury-Vapor Apparatus.
Tests for the Efficiency of the Eye under different Systems of Illumina-
tion, and a Preliminary Study of the Causes of Discomfort.
Street Lighting with Ornamental Luminous Arc Lamps.
Some Phases of the Illumination of Interiors.
Illumination and Eye Strain.
Home Illumination.
A Photometer Screen for use in Tests of Street Illumination.
The Flame Carbon Arc Lamp.
The Illumination of Motion Picture Projectors.
Street Lighting of Greater New York.
The Illumination of Passenger Cars.
Some Home Experiments in Illumination from Large Area Light
Sources.
Gas Lighting in an Exhibition Hall.
Metal Reflectors for Industrial Lighting.
Vision as Influenced by the Brightness of Surroundings.
A Practical Solution of the Problem of Heterochromatic Photometry.
ANNUAL REPORT OF THE GENERAL SECRETARY 689
Nearly all the papers and many of their attending discussions
have laid more or less stress upon the hygienic aspects of light
and illumination. Greatest interest has been manifest in those
papers dealing with the commercial application of the scientific
principles to lighting practise, say in the design or re-design of
installations.
Not any of the papers presented at the 191 3 convention are
included in the foregoing list.
Beginning with the first of the year the Transactions was
published in a new form. The general make-up of the publica-
tion has been improved. Paper free from glare has been used
throughout. The new style has increased the cost of publi-
cation between 25 and 30 per cent.
committies.
Some twenty-five permanent and temporary committees have
conducted the work of the society with rather unusual activity.
A brief record of the success of their work is given in the follow-
ing paragraphs.
1913 Convention Committee. — The 191 3 convention — thanks
to the good work of the committee — surpassed in general
excellence any previous convention. The attendance of out-of-
town delegates was considerably larger than that of any previous
year. An excellent program of papers, well balanced with com-
mercial and scientific subjects, lively discussions, and a generous
complement of amusement combined to make the convention an
unusual success.
Committee on the Glore from Reflecting Surfaces. — This com-
mittee followed up the work which it started in 1912. It has
confined its efforts chiefly to giving publicity to the evils arising
from the glare of reflected surfaces, especially from glazed
paper, and to the collection of data and information pertaining
to the use of unglazed paper. An eight-page leaflet entitled
"Glare" was published and circulated. The leaflet consisted of
glazed and unglazed paper and was designed primarily for the
purpose of emphasizing the evils of glare from paper. The com-
mittee has communicated with a number of leading publishers of
69O TRANSACTIONS I. E. S. — PART II
books and periodicals. Several publishers have already adopted
unglazed paper through the efforts of the committee. In the
opinion of the committee glazed paper will be eliminated from
general use as soon as this change can be effected without an
unreasonable increase in cost. With an increased demand for
unglazed paper, it is very likely that an entirely satisfactory paper
will be forthcoming.
Committee on Research. — The committee has drafted some ex-
cellent plans for its procedure. If these are pursued even to a
small degree in the future, noteworthy results must accrue.
Briefly the committee has planed to make its services (1) co-
operative and advisory, and (2) initiatory. The committee favors
co-operation with several committees of the Society and sister
societies, with departments of universities, research laboratories
and individuals, when matters of research are concerned. The
committee proposes (a) to suggest research problems to parties
desirous and competent to do valuable work, (b) to compile a
list of subjects in which research seems to be advisable, and
(c) to arrange a bibliography on the various phases of illumina-
tion. While the fulfillment of these functions and plans has not
progressed sufficiently to warrant definite achievements, results
may be expected in the future.
Committee on Collegiate Education. — The work of this com-
mittee has been conducted in accordance with the following plan ;
(1) to determine which colleges and universities are giving any
attention to illuminating engineering; (2) to find out exactly
the character and extent of such work as is being done; (3) to
call attention of college authorities to the importance of greater
and adequate instruction in all matters pertaining to the use of
light, with a view to (a) bringing about the introduction of
courses in illumination, and (b) looking to the extension and
amplification of present courses. To ascertain the status of il-
luminating engineering in the curricula of colleges, the committee
has sent a letter to a number of college presidents, asking for
information, in regard to courses in illumination in their insti-
tutions. A sufficent number of returns have not yet been re-
ceived from these inquiries to warrant a statement at this time..
ANNUAL REPORT OF THE GENERAL SECRETARY 691
Committee on Reciprocal Relations with other Societies. —
The excellent work of this committee, which was appointed
for the first time in 1912, has been continued during the present
year. The committee has been in touch with some twenty or
thirty societies and has arranged joint meetings with several of
them, notably the American Gas Institute, The National Com-
mercial Gas Association, The National Electric Light Associa-
tion, The American Academy of Medicine, the School Hygiene
Congress and the American Medical Association. In acquaint-
ing other organizations with the objects and province of the
Illuminating Engineering Society this committee has rendered
a service of great value.
Committee on Sustaining Membership. — Through the efforts
of this committee the companies whose names appear on a pre-
vious page all became sustaining members within a period of five
months. The revenue from sustaining membership has increased
the funds of the society so that the work on a more enlarged and
efficient scale might be started.
Committee on Nomenclature and Standards. — The committee
has held several meetings, and has been in touch with the stan-
dards committees of several American and foreign societies,
which are interested in the science and art of illumination. It
has formulated plans for aiding the formation of an international
photometric commission. It has also proposed a number of re-
vised definitions of standards and of terminology, which have
been transmitted to the International Photometric Commission
and interested societies.
Committee of Factory Lighting Legislation. — Appointed this
year for the first time to consider matters of legislation where il-
lumination is concerned, this committee has undertaken and ren-
dered valuable service in a field which heretofore has received
scant and insufficient attention. Upon invitation, its members
attended a meeting of the New York State Factory Commission
and made a number of recommendations relating to the lighting
of factories and workrooms, with special reference to adequacy
and quality of illumination. A bill incorporating these recom-
692 TRANSACTIONS I. E. S. PART II
mendations was submitted to the Legislature and became a law
in New York State on October 1, 1913. The committee was
also represented at a conference of the Heights of Buildings
Committee of the Board of Estimate and Apportionment of the
City of New York, at which the illumination of interiors by day-
light and by artificial light was discussed.
Section Development Committee. — During the year the com-
mittee completed work on the preparation of a guide on section
management which had been started in the previous year. The
guide constitutes a convenient arrangement of notes and sug-
gestions which tells in a general way how the affairs of section
may be advantageously conducted. It is believed that it will go
a long way toward promoting co-operation, co-ordination and
efficiency in the work of the society. Heretofore new boards of
managers have had to undertake the management of sections
with a very indefinite knowledge of the duties and responsibilities
to be discharged. Now any board, though it be entirely new,
may in an hour or two become familiar with what has been
found to be good procedure and what work was done by the pre-
vious board. Hereafter it is likely that the preliminary plans
and details incidental to the beginning of each season will be
arranged with greater dispatch. Copies of the guide have been
issued to the members of all the section boards as well as to
a number of officers of the society.
Committee on Finance. — The Committee on Finance has exer-
cised general supervision over the affairs of the Society. It has
approved all appropriations and disbursements.
Committee on Illumination Primer. — A few slight revisions
were made in the first edition of the primer which was published
last fall. Of this revised edition approximately a quarter of a
million copies have been printed. Over 220,000 copies of the
latter edition were sold to lighting companies throughout the
country for distribution to their customers. The primer has been
widely advertised and commended in trade, and other pournals
both in this country and abroad. No work of the Society has
met with greater approval on the part of the public and the light-
ing industry in general. Certainly no publication of its kind has
ANNUAL REPORT OF THE GENERAL SECRETARY 693
done more to spread the gospel of good illumination, particularly
in the way of creating a desire for better standards of illumina-
tion.
Committee on Papers. — The Committee on Papers has passed
upon all papers submitted to the Society within the past year.
It has earned the distinction of having arranged for the 191 3
convention a program which has been pronounced the best bal-
anced set of papers ever presented before a meeting of the
Society.
Committee on Editing and Publication. — While practically all
the publication work of the Society has been done in the general
offices, the Editing Committee has served in an advisory capacity
and passed upon from time to time questions involving the edi-
torial policy of the Society. A guide for authors, setting forth
the editorial requirements and style of papers has been prepared
for distribution. The committee believes that this pamphlet will
promote a degree of desirable uniformity in all papers and at the
same time obviate considerable expense in publication.
Committee on Advertising. — In accordance with instructions
from the Council, the committee has made no special attempt to in-
crease the amount of advertising contracted for during the year
1912. The net income derived from advertising, $1,097.14, was
relatively, a little higher than that for a corresponding period of
1912.
Committee on Membership. — No active general campaign for
increasing membership has been conducted. Through the chair-
man of the various section Membership Committees, a conserva-
tive effort was made to bring into the Society those members
who are most likely to contribute to and profit by its work. Dur-
the year the total membership was increased by 62 members.
One hundred and forty-eight applications and reinstatements
were received largely as a result of the committee's efforts.
Committee on Progress. — The work which this committee did
is indicated by the excellent and comprehensive report of progress
in the field of illuminating engineering during the past year,
which was submitted at the convention. The report* sets a high
standard of comparison for all future progress reports.
* Trans. I. E. S., Vol. VIII, No. 7 (October, 1913), p. 323-
694 TRANSACTIONS I. E. S. PART II
Committee on Joint Session with the American Gas Institute. —
This committee was appointed solely for the purpose of arrang-
ing an illuminating engineering session at the annual meeting of
the American Gas Institute which was held in Richmond in
October, 1913. For this session the committee arranged for a
number of papers and talks on various phases of illuminating
engineering and the work of the Society itself.
Committee on Joint Meeting with the International Congress
of School Hygiene. — A session at which several papers on the
hygienic aspects of light and illumination were presented was
arranged by the committee. The Congress manifested a great
deal of interest in this particular session.
Committee on 1915. — This committee was appointed for the
purpose of considering the advisability of holding the 191 5
convention of the Society in San Francisco during the Panama
Pacific Exposition. No definite recommendations were made be-
cause the committee believed the time of that convention was
too far off.
Committee on Popular Lectures. — The Committee on Popular
Lectures was appointed at the beginning of the year to make
available popular lectures on good lighting practice. With such
lectures it is hoped that it will be possible to create on the part
of the public a better appreciation of good lighting and ultimately
raise the standards of illumination. While it is planned to cover
other fields of lighting, the first efforts will be confined to resi-
dence, store, industrial lighting and office lighting. Sub-com-
mittees have been appointed — one for each of these four classes
of lighting, the members of which are experts familiar with the
best practice. In order that lectures may be useful, without
evincing commercial prejudice, a special endeavor has been made
to balance the committees with reference to the interests repre-
sented that they may be free from any commercial bias. It is
proposed that each of these sub-committees shall prepare one or
more lectures on the subjects assigned. Photographs from which
lantern slides may be made will be collected to illustrate by
example what is considered good lighting practise, and on ihe
ANNUAL REPORT OF THE GENERAL SECRETARY 695
other hand to emphasize the evils of poor illumination. A careful
selection of the committees has just been completed and it is
planned to have available within the next month or two at the
general offices of the Society, a series of lectures which may be
loaned to societies and associations, lighting companies, both gas
and electric, and other organizations which are interested in
illumination.
To the Committee of Tellers which counted with patience and
care the ballots of the annual election, and to the Committee on
the Formation of a Lake Erie Section, which considered and re-
ported the possibilities of such a section, special thanks is due.
The latter proposal while encouraging was left for future de-
velopment.
Delegates. — During the year a number of representatives have
been appointed, upon invitation, to boards and committees of
other organizations, (International Gas Congress, Committee on
Organization International Electrical Congress, American Asso-
ciation for the Conservation of Vision, United States National
Committee of the International Commission on Illumination) in
order that the Society might keep in touch with work with which
it is concerned. The services of these representatives deserves
the thanks of the Society.
general.
Here, at the end of another year, there is cause to look back
with grateful appreciation upon praiseworthy efforts of the
officers, committeemen, section boards, and many other members
who have given unstintingly of their time and thought to the
work and objects which the Society seeks to promote. The
measure of success attained must be ascribed entirely to their
work.
Worthy of record, also, is the encouraging co-operation which
has been obtained from sister organizations in promoting what
has been aptly called the gospel of better illumination. There
have been numerous evidences of increasing and greater respect
for illuminating engineering or the science and art of illumina-
tion as many prefer to call it. Bonds of common interest be-
tween the I. E. S. and many societies have been established and
strengthened.
696 TRANSACTIONS I. E. S. — PART II
Further, the initial activities of the German Illuminating En-
gineering Society, and the promise of organization of a French
society before long, have attracted world-wide attention. In a
word, illuminating engineering is gradually coming into its own.
Let those who wish predict when illuminating engineering will
take its place among the firmly established professions. For the
present it is gratifying enough to observe the rapid advancement
in that direction.
Some reason there might be for referring in this report to the
generally consistent progress, the notable developments, and the
numerous examples of scientifically designed installations, in the
lighting field during the past year if all these were not already
chronicled elsewhere.
Respectfully submitted,
Joseph D. Israel,
General Secretary.
TP
700
133
v. 8
Illuminating engineering
^frifili ' ~
Engineering
PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET
UNIVERSITY OF TORONTO LIBRARY
ENGIN
STORAGE